Skip to main content

Full text of "The journal of the Polynesian Society"

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 



at |http : //books . google . com/ 




: rSltXS VfiiGHt OtrNNlNC ^1 



BEQUEST 
UNlVTiRSlTY oi^MICHIGANi 1' 

GENERAL LIBRARY 

V^^^^ : __ _; 



M 



'V-^<>.v..u 



Digitized by 



Goog 



GrN 

.?7? 



Digitized by 



Google 



Digitized by 



Google 



THE JOURNAL 



OF THE 



POLYNESIAN SOCIETY 

CONTAINING 

THE TRANSACTIONS AND PROCEEDINGS 
OF THE SOCIETY 



"v^on,. :K.'xrvxxx. 



1919. 



PHINTBD FOR THE SOCIETY BY ThOMAS AvEEY, DbVON StEEET. 



1919. 



Digitized by 



Google 



Digitized by 



Google 






CONTENTS OF VOL. XXVIII. 



No. 109.— Maech, 1919. 



Annual Meeting of Society . . 

Annual Report of Council 

Balance Sheet 

Members of the Society 

List of Exchanges 

Books, etc. received-in exchange 

The Land of Tara. By Elsdon Best 

The- Fatherland of the Polynesians. By S. Percy Smith 

Traditions of and Notes on the Paumotu Islands. By the Rev. P^re H 

Audran 
The San Cristoval Heo. By Rev. C. E. Fox 
Traditions and Legends of Southland. By H. Beattie 
The Science Congress, Christchuroh, N.Z. 
Notes and Queries — The Teaching of Ethnology . . 
Proceedings . . 



PAGET 

i 

i 

iv 

V 
X 

xii 

1 
18 

31 
39 
42 
52 
54 
64 



No. 110.— June, 1919. 

History and Traditions of Rarotonga. By Te Ariki-tara-are . . 

The Land of Tara. By Elwdon Best 

The Great Mum. By a Taranaki Veteran 

Further Notes on the Heo of the Solomon Islands. By the Rev. C. E. Fox 

Very like Scalp-taking. By S. Percy Smith 

Rangi-hua-moa. By Geo. Graham 

An Account of Kupe and Tainui. By Geo. Graham 

The Visit of Dentrecasteaux to the North Cape, New Zealand . . 

Notes and Queries — Malay o- Polynesian 

The Pacific Story of Jonah and the Whale 
Proceedings 



55^ 
79 
97 
103, 
106 
107 
111 
117 
121 
121 
122 



No. 111. — September, 1919. 



By Te Ariki-tara-are 
By H. Beattie 



The Land of Tara. By Elsdon Best 

History and Traditions of Rarotonga. 

Traditions and Legends of Southland. 

An Ancient Carved Fare 

Traditions of and Notes on the Paumotu Islands 

Polynesian Lingfuistics. By Sidney H. Ray 

Gilbert's Account of Easter Island. By H. D. Skinner 

Notes and Queries — The Wanaka District 

Proceedings 



123 
134 
162 
16a 
161 
168 
178 
180" 
181 



Digitized by 



Google 



CONTENTS. 



No. 112. — Decembeb, 1919. 

History and Traditions of Rarotonga. Part VII, By Te Aj-iki-tara-are 

The Paumotu Conception of the Heavens and of Oreation. By J. L. Young 

Traditions and Legends of Southland. By H. Beattie 

Nine Folk-lore. By G. N. Morris . . 

The Maori Belief in the Supernatural Powers of certain Axes . . 

Traditions of and Notes on the Paumotu Islands. By the Rev. P^re H 

Audran 
A Maori Stone Axe. By H. D. Skinner 
Notes and Queries — ^The Three Fingers in Maori Carvings 

Polynesian Voyages 
Proceedings 



PAGE 

183 
209 
212 
226 
229 

232 
240 
242 
242 
243 



Digitized by 



Google 



VOL. XXVIII.— 1919. 

ANNUAL MEETING OF THE SOCIETY, 
For the Year Endino 3 Ist December, 1918. 



Thb Meeting took place on the 29tli January, 1919, at the Hempton Room, 
-where the Society's Library is installed. There were several members present, 
the President being in the chair. 

The minutes of the last annual meeting were read and confirmed, and then the 
Annual Report of the Council and the year's accounts, which were passed and 
ordered to be printed in the March number of the * Journal.' They will be found 
below. 

Ill Accordance with the rules, the President retired and two members of the 
Council also, the latter by ballot. Mr. S. Percy Smith was re-elected President, 
and Messrs. M. Fraser and W. W. Smith re-elected to the Council. 

A vote of thanks was paased to Mr. W. D. Webster for auditing the accounts, 
and he consented to act in the same capacity this ensuing year. Also a vote of 
thanks was accorded Mr. W. H. Skinner for preparing the Index to last year's 
volume. 

The following new members were then elected : — 
Hubert E. Vaile, Queeu Street, Auckland. 

The Adalbert College, Western Reserve University, Cleveland, Ohio, 
•U.S.A. 

A discussion took place on a paragraph in the annual report referring to the 
reading of papers before the Society, and the question was finally referred to the 
Council. 

A resolution was passed expressing the opinion of the members present as to 
the desirability of the appointment of a Curator of the New Borough Museum, 
New Plymouth, which was ordered to be conveyed to His Worship the Mayor. 



ANNUAL EEPORT OF THE COUNCIL, 
For the Year Ending 31st December, 1918 



Thb Council of the Society has pleasure in presenting to the Annual Meeting its 
Twenty-sixth Report, showing in brief form the activities of the Society during 
the past twelve mouths. 

The most noticeable feature of the past year has been the removal of the 
headquarters into the commodious apartments of the Hempton room where our 
library and office have been installed, and which was opened by His Worship the 
Mayor, as referred to in our last annual report. The additional accommodation 
thus provided by the generosity of a lady of New Plymouth, has allowed- 
of our library being for the first time arranged in a form where access to 
the books has become easy, just as in other libraries ; whilst the comfort in 
working by the officers of the Society has been greatly appreciated. The books 



Digitized by 



Google 



ii JOURNAL OF THE POLYNESIAN SOCIETY. 

that have accumulated by exchange and gift during the past twenty-seven years 
now occupy 240 feet of shelving, with walJ space for at least three times as 
much. Arrangements have been made for opening the library to the public on 
Monday and Saturday afternoons, and on Wednesday eyening^ for the purpose 
of reference or study, and for the issue of books to members, an officer of the 
Society being in attendance at these times. We should be glad to welcome more 
visitors to the library for purposes of study, for though our collection naturally 
consists mostly of works on ethnology, the liberality of our exchanges provides us 
with works on most of the sciences ; our American exchanges are particularly 
liberal in this respect. 

The rules passed by the first meeting of the Society in 1892, defined our 
objects as follows : ** The= Society is formed to promote the study of the anthro- 
pology, ethnology, philology, history and antiquities of the Polynesian races by 
the publication of an official journal to be called * The Journal of the Polynesian 
Society,' and by the collection of books, manuscripts, photographs, relics, and 
other illustrations.'* We think that these objects have been fairly well carried 
out as exemplified in the 27 volumes of our journal, which contain original matter 
that would otherwise, but for our Society, have been lost for ever to the world and 
the future students of the race. Naturally the great problem that presents itself 
for solution is the origin of our Pacific Island peoples. The mystery surrounding 
this quesl^on has had for very many people a great fascination. The tendency of 
all observation so far is to carry the ancestors of the people back to Indonesia, and 
this view is supported by all lines of study. Beyond that, the matter is in some 
doubt, though some of us hold strong opinions about it and see in the early Aryan 
migrants into India the ancestors of the Polynesians. Towards the final solution 
of the question we may fairly claim to have helped to lay the foundations, by ^e 
original matter preserved in our journal. But much yet remains to be done. For 
instance, no reasonable hypothesis has as yet been formidated accounting for the 
Easter Island statues, some light upon which may, however, be thrown by the 
publication of Mrs. Scoresby Routledge's forthcoming work, the results of a year's 
«ojoum on that island. 

Our Rules, in defining the functions of the Society, do not provide for the 
reading and discussion of papers by members at meetings to be held for the 
purpose, as is the custom in similar societies. But there is no reason why this 
course should not be adopted if members wish it. Such discussions often bring out 
points that are not alluded to in the papers read, and thus are beneficial. Were 
this course adopted, aided by occasional lectures — not necessarily on strict ethno- 
logical lines— it might tend to popularise the work of the Society, and if the public 
generally were invited to attend, the result would furnish people with something 
to think about in addition to their daily round of duties. 

Our membership was slightly increased during the year, and on December 
31st the roll was as follows: — 

Patrons . . . . . . 3 

Honorary Members 12 

Corresponding Members . . 13 

Ordinary Members . . 158 

Total ..186 

• During the year three members resigned and four died — Mr. W. Kerr, S.M., who 
was a member of our council for several years; Dr. H. CJoUey March, an original 
member, and a well-known writer on ethnology; Mr. A. H. Tumbull, of Welling- 
ton, weU-known in connection with his celebrated library of works relating to New 



Digitized by 



Google 



ANNUAL REPORT OF COUNCIL. iK 

Zealand; and Mr. J. H. WIIhou, judge of the Native Land Court, who succumbed 
to the influenza epidemic. We have also reason to fear that another promising 
member whs killed in J'rance, but as yet have no certain information. The number 
of members who joined the Society at its formation in 1892 are now reduced to 17, 
We exchange publications with 44 societies, universities,, Goveriiment departments, 
etc., and receive in exchange more than an equivalent from some of them. 

There is a considerable demand for sets of our publications from various parts 
of the world, which we are no longer able to meet so far as some of the earlier 
volumes of our ** Journal" are concerned, as they are out of print. If satisfactory 
arrangements can be made these ought to be reprinted; it would pay the Society in 
the long run to do so. Furthermore, our library is increasing so fast that 
immediate expansion of the shelving accommodation has become necessary. 

On December <31st there were 27 members iu arrear for one year, nine for two 
years, and four for three years. These latter will have to be struck off the roll in 
accoi^auce with our rules. 

It will be seen from the treasurer's statement that financially we are in a good 
position, as there is a balance in hand of £45.~ Our correspondence wi£h all' parts 
of the world increases, for the Society is g^etting known far and wide, so that we 
are frequently applied to for information on Polynesian ethnological subjects. 



Digitized by 



Google 



BALANCE SHEET. 



0<0000t^000000o^ 






M S 

S S 
- I 

p rt 

Ph O 

W 

» 
o 

pq 



3 .MMM 



*5i 

j 

^ o 



g ^ ^- ^- ^• 

.2 "^ ^ o^ CO 
Ph^ pod 

S5;z; 



'M 



< 



■pjg i;. C ^4 

Hi ft 

rH ^ ^ CO 

I— I Q -M 



-5 a S;s 
5 2J3 § 



-3 

15 






5-*^ 
P0Q<O 






i 5 3-3 
(Pnpqp- 






s 

OS 

00 











o t^ 


t>- 




© 00 


00 






r^ 




Ok 00 


r«-. 




^ 2 


o» 






«3 




OS 












o> 






1 '^ 






:&4 






II' 






II 






•Si 






-S 






gft- 






^ ^ 






§ s 






<1^ 






tn ■*^ 






2 « 




g 


il 




D 


^1 




^ 


►» - 




Q 


pq ^ 




< 






< 


^ ^ O 00 


t^ 


. -<*< O CO 


00 


g 


tc ^ 




,,OS pHt^ 


t^ 


<1 


«4?Q0 


Oi 


o 


To Balance let January, 1918 
, , Donation . . 
,, Interest 


53 



a 



W 55 
jz; 



^ 






pf 
g 

I f=° 

is 

80 
IS 



^ 



Digitized by 



Google 



VOL. XXVIII.— 1919. 
MEMBERS OF THE POLYNESIAN SOCIETY 

As FROM IsT January, 1919. 



The sign * before a name indicates an oiiginal member or founder. 

4> this list will be published annually, the Secretaries would be obliged if members will 

supply any omission, or notify change of address. 



PATRONS: 

The Eight Hon. Baron Plunket, K.C.M.G., K.C.V.O., ex-Govemor of New 
Zealand, Old Oomianght, Bray, County Wicklow, Ireland. 

The Eight Hon. Baron Islington, K.C.M.G., D.S.O., ex-Govemor of New 
Zealand, Government Offices, Downing Street, London. 

His Excellency The Eight Hon. The Earl of Liverpool, M.V.O., G.O.M.G., 
Governor G^eneral of New Zealand. 

HONORARY MEMBERS. 

Eev. E. H. Codrington, D.D., Chichester, Enghind. 

Eev. Prof. A. H. Sayce, M.A., Queen's College, Oxford, England 

Eight Hon. Sir J. G. Ward, Bart., K.C.M.G., P.C, LL.D., M.P., WeUington 

H. G. Seth-Smith, M.A., c/o W. T. Williams, 7, St. Helens Place, Bishopsgate 

Street, London, E.C. 
Prof. Sir W. Baldwin Spencer, M.A., C.M.G., F.R.S., The University, Melbourne 
*Edward Tregear, I.S.6., Wellington 

Dr. A. Haddon, M.A., D.Sc, F.E.S., 3, Cranmer Eoad, Cambridge, England 
Churchill, W., B.A., F.E.A.I., Yale Club, 30, West Forty-fourth Street, New 

York 
Sir J. Q. Eraser, D.C.L.. LL.D., Litt. D., Brick Court, Middle Temple, London, 

E.C. 
♦Etsdon Best, Dominion Muneum, Wellington 

Chas. M. Woodford, C.M.G., The Grinstead, Partridge Green, Sussex, England 
S. H. Eay, M.A., F.E.A.I., 218, Balfour Eoad, Dford, Sussex^ England 

CORRESPONDING MEMBERS: 

Eev. T. G. Hammond, Putaruru, Auckland 

Major J. T. Large, Masonic Institute, H. M. Arcade, Auckland. 

Hare Hongfi, 3, Stirling Street, Wellington 

Tati Salmon, Papeete, Tahiti 

Tunui-a-rangi, Major H. P., Carterton 

Whatahoro, H. T., Putiki, Whauganui 

Christian, P. W., Otaki 

The Eev. 0. E. Fox, San Cristoval ; vi^ Ugi, Solomon Islands 

Skinner, H. D., B.A., D.C.M., Hocken Library, Dunedin 

Eev. P^re Herve Audran, Faknhiva, Tuamotu, Tahiti 

M. Julien, His Excellency, Governor of French Oceania, Tahiti 

A. Leverd, Papeete, Tahiti 



Digitized by 



Google 



vi JOURKAL OF THE POLYS'ESIAN' SOCIETY. 

ORDINARY MKMBKR8: 

1894 Aidred, W. A., Bank of New Zealand, Wellingtoa 

1899 Atkioson, W. E., Whacgauui 

1916 Avery, Tho8., New Plymouth 

1918 Adalbert College, Weeteru Reserve Univernty, Cleveland Ohio, U. S. A. 

1892 ♦Birch, W. J. Thoresby, Marton 

1892 ♦Barron, A., Macdonald Terrnoe, Wellington 

1894 Bamford, £., Amey Road, Aackbind 

1896 British and Foreign Bible Society, 146, Queen Victoria Street, London, B.C. 

1898 Buchanan, Sir W. C, Tupurupuru, Masterton 

1902 Boston City Library, Boston, Mass., U.S.A. 

1907 Buick, T. Lindsay, F. R. Hist.S., Press Association, Wellington 

1917 Brown, Prof. J. MacMillan, M.A., LL.D., Holmbank, Cashmere Hills, 

ChriHtchorch 

1909 Bollard, G. H., Chief Surveyor, New Plymouth 

1910 Burnet, J. H., Virginian Homestead, St. Juhn's Hill, Whanganui 

1910 Burgess, C. H., New Plymouth 

1911 Bird, W. W., Inspector Native Schools, Napier 

1913 Buddie, R., H. M. S. *' Northampton,'* c/o General Post Office, London 

1914 Brooking, W. F., Powderham Street, New Plymouth 
1914 Beattie, Herries, P. O. Box 40, Gore 

1916 Bottrell, C. G., High School, New Plymouth 

1918 Beyers, H. Otley, Professor Department of Anthropology, University of the 

Philippines, Manilla 

1918 Brown, A. Radcliffe, Nukualofa, Tonga Island 

1892 ♦Chapman, The Hon. F. R., Wellington 

1892 Chambers, W. K., Fujiya, Mount Smart, Penrose, Auckland 

1893 Carter, H. C, 475, West 143rd Street, N.Y. 

1894 Chapman, M., Wellington 

1896 Cooper, The Hon. Theo., Wellington 

1900 Cooke, J. P., c/o Alexander and Baldwin, Honolulu 

1903 Chatterton, Rev. F. W., Te Rau, Gisbome * 
1903 Cole, Ven. Archdeacon R. H., D.C.L., Pamell, Auckland 

1908 Coughlan, W. N., Omaio, Opotiki 
1908 Carnegie Public Library, Dunediu 
1908 Carnegie Public Library, New Plymouth 
1910 Cock R., New Plymouth 

1917 Cowley, Matt, P. O. Box 72, Auckland 

1918 Chambers, Bernard, Te Mata, Havelock North 
1918 Comey, Geo., Devon Street, New Plymouth 

1918 Crooke, Alfred, S.M., Devon Street, New Plymouth 

1892 ♦Denniston, The Hon. Sir J. E., Christchurch 

1902 Dulau & Co., 38, Soho Square, London 

1902 Drummond, Jas., ** Lyttelton Times " Office, Christchurch 

1903 Dixon, Roland B., Ph.D., Harvard University, Cambridge, Mass , U.S.A. 

1910 Downes, T. W., P. O. Box 119, Whanganui 

1911 Drew, C. H., New Plymouth 

1917 Dominion Museum, Wellington 

1918 Davidson, J. C, Motunui, Waitara 



Digitized by 



Google 



MEMBERS OF SOCIETY, vii 

1892 ♦Emernon, J. 8., 802, Spencer Street, Honolulu, Ha^raiian Islandii 
1894 Ewen, C. A., Commercial Union Insurance Co., Wellington 

1918 Etheridge, Robt. Director, Austniliau Museum, Sydney 

1896 Fletcher, Rev. H. J., Taupo 

1900 Forbee, B. J., 5, Hamilton Street, Sydney, N. S. W. 

1901 Firth, John F., Surrey Office, Nelson 

1902 Fnuer, M., New Plymouth 

1902 Fisher, T. W., Tikitiki Road, Te Mapara, Te Kuiti 

1903 Fowlds, Hon. G., Auckland 

1906 Held Museum of Natural History, The, Chicago, U.S.A. 

1912 Fisher, Mrs. Lillian S., 560, Hancock Street, Brooklyn, New York, U.S.A. 

1912 Fisher, F. Owen, c/o Credit Lyonaise, Biarritz, B.P., France 

1913 Fildes, H., Box 740, Chief Post Office, Wellington 

1892 •Gudgeon, lieut.-Col. W. B., C.M.G., 39, King's Parade, Devonport, 

Auckland 

1902 Gill, W. H. Marunouchi, Tokio, Japan 

1902 Graham, Geo., c/o Commercial Union, P.O. Box 166, Auckland 

1910 Godding, Fred W., U.S. Consul G^eneral, Guayaquil, Ecuador 

1898 Hastie, Miss J. A., 11, Ashbum Place, Cromwell Road, London 

1908 Hallen, Dr. A. H. Clevedon, Auckland 

1909 Holdflworth, John, Swarthmoor, Havelock, Hawkes Bay 

1910 Hawkes Bay Philosophical Society, P.O. Box 166, Napier 
1910 Hooken, Mrs. T. M., Hocken Library, Dunedin 

1910 Home, Dr. G^rg^, New Plymouth 

1911 Heimbrod, G., F.R.A.I., Lautoka, Fiji 
1911 Henniger, Julius, Motuihi Island, Auckland 

1917 Hocken Library, Dunedin 

1918 Hodgson, N.V., c/o Norman Potts, Opotiki 

1918 Harrie, Rer. G. F., The Vicarage. Vivian Street, New Plymouth 

1918 Hart, Henry H., 3751, CUiy Street, San Francisco 

1907 Institute, The Auckland, Museum, Auckland 
1907 Institute, The Otago, Dunedin 

1892 'Johnson, H. Dunbar, Judge N.L.C., 151, Newton Road, Auckland 

1918 Johnston, E. G., Education Board Office, New Plymouth 

1902 Kelly, Thomas, New Plymouth 

1910 King, Newton, Brooklands, New Plymouth 

1894 Lambert, H. A. Belmont, Tayforth, Whangauui 

1911 Lysnar, W. D., Gisbome 
1913 List, T. C, New Plymouth 

1913 Lysons, E. W. M., New Plymouth 

1916 Leathani, H. B., M.R.C.S., L.R.C.P., Lond., New Plymouth 

1917 Ledingham, T. J., ** Montecute," St. KUda, Melbourne 

1917 List, C. S., Rata Street, Inglewood 

1918 Laughton, Rev. J. G., Ruatahuna, vi& Rotorua 

1892 *Mar8ha]l, W. S., Maungaraupi, Rata 

1892 'Major, C. E., 22, Empire Buildings, Swanson Street, Auckland 



Digitized by 



Google 



viu JOURNAL OF THE POLYNESIAN SOCIETY. 

1897 MarshaU, J. W., Tututotara, Marton 

1897 Marshall, H. H., Motu-kowhai, Marton 

1907 Minister for Internal Affairs, The Hon., Wellington 

1912 Marsden, J. W., Isel, Stoke, Nelson 

1915 Mahonej, B. G., c/o G. Mahoney, Esq., Ruatoki, Taneatua 

1916 Mitchell, Library, The, Sydney 

1917 Marshall, P., M.A., D.Sc, F.G.S., Collegiate School, Whanganni 

1918 McDonell, A. F., Queen Street, Auckland 

1918 Morris, G. N., Resident Commissioner, Nine Island 

1918 Missionary Research Library, 25 Maddisoh Arenue, New York 

1895 Ngata, A. T., M.A., M.P., Parliamentary Buildings, Wellington 
1900 Newman, W. L., New Plymouth 

1902 New York Public Library, Astor Buildings, New York 

1906 Newman, Dr. A. K., Hobson Street, Wellington 

1894 Partmgton, J. Edge, F.R.G.8., Wyngates, Burke's Road, Beaconsfield, 
England. 

1907 Public Library, Auckland 
1907 PubUc Library, Wellington 
1907 Public Library, Sydney, N.S.W. 

1907 Philosophical Institute, The, Christchurch 

1907 Postmaster General, ITie Hon. The, Wellington 

1913 Potts, Norman, Opotiki 

1914 Parliamentary Library, (the Commonwealth), Melbourne 
1917 Patuki, J., Topi, M.L.C., Ruapuke Island, Invercargill 

1917 Platts, F. W., Resident Commissioner, Rarotonga Island 

1892 ♦Roy, R. B., Taita, Wellington 

1903 Roy, J. B., New Plymouth 

1918 Rylands, John, Library Deansgate, Manchester University, England 
1918 Rockel, R. H., M.A., Gover Street, New Plymouth 

1892 ♦Smith, W. W., F.E.S., Pukekura Park, New Plymouth 

1892 ♦Smith, F. S., Blenheim 

1892 *Smith, M. C, Survey Department, Wellington 

1892 ♦Smith, S. Percy, F:R.G.S,, New Plymouth 

1892 'Stout, Hon. Sir R., K C.M.G., Chief Justice, Wellington 

1892 ♦Skinner, W. H., Chief Surveyor, Christchurch 

1896 Smith, Hon. W. O., Honolulu, Hawaiian Islands 

1904 Smith, H., Guthrie, Tutira, via Napier 

1904 Samuel, The Hon. Oliver, M.L.C., New Plymouth 

1905 Schultz, Dr. Erich von, late Imperial Chief Justice, Samoa, Motuihi Island, 

Auckland 
1907 Secretary of Education, Wellington 
1910 Savage, S., Rarotonga Island 

1915 Smith, Alex., c/o W. W. Smith, New Plymouth 

1916 Shalfoon, G., Opotiki 

1913 Tribe, F. C, Vogeltown, New Plymouth 

1915 Thomson, Dr. Allan, M.A., D.Sc., F.G.S., A.O.S.M., Museum, Wellington 

1916 Te Anga, Hone Tukere, N.L. Court Office, Whanganni 

1917 Tarr, W., Gk)vernment Printing Office, Nukualofa, Tonga Islands 

1918 Trimble, Harold, Inglewood 



Digitized by 



Google 



MEMBERS OF SOCIETY, 

1911 Vibaud, Rer. J. M., Hiruharama, Whanganui 
1919 Vaile, Hubert E., Queeo Street, Auckland 

1892 «WilUam8, Archdeaooo H. W., Gisboioe 

1894 Wilson, A., Hangatiki, Auckland 

1896 Williams, F. W., Napier 

1896 Wilcox, Hon. G. A., Kauai, Hawaiian Islands 

1898 Whitney, James L., Public Library, Dartmouth, Boston, U.S.A. 

1902 Webster, W. D., New Plymouth 

1903 Walker, Ernest A., M.D., New Plymouth 
1910 Wilson, Sir J. G., Bulls 

1912 Westervelt, Rev. W. D., Honolulu, Hawaiian Islands 

1914 Waller, Captain W., Moturon, New Plymouth 

1915 Williams, H. B., Turihaua, Gisbome 

1915 Wilson, Thos., Captain, New Plymouth 

1916 Welsh, R. D., Hawera 

1916 White, Percy J. H., New Plymouth 

1917 Wheeler, W. J., Inspecting Surveyor, Gisbome 

1917 Wilkinson, C. A., M.P., Eltham 

1918 Wallace, D. B., Masonic Club, H.M. Arcade, Auckland 
1918 Western, T. D., Puketapu, Bell Block, New Plymouth 

1918 Wilson, Kenneth, M.A., 92, Rang^tikei Street, Palmerston North 

1892 *Young, J. L., c/o Henderson and Macfarlane, Auckland 



PRESIDENTS— Past and Present : 

1892-1894— H. G. Seth-Smith, M.A. 

1895-1896— Right Rev. W. L. Williams, M.A., D.D. 

1896-1898— The Rev. W. T. Habens, B.A. 

1901-1903— E. Tregear, I.S.O., etc. 

1904-1919— S. Percy Smith, F.R.G.8. 



Digitized by 



Google 



LIST OF EXCHANGES. 



THE following is the List of Societies, etc., etc., to which the Journal is sent, 
and from most of which we receive exchanges : — 

Anthropologic, Soci6t^ d', 15 Rue Ecole de Medicin, Paris 

Anthropologia, Societa, Museo Nazionale di Anthropologia, Via Gino Gapponi, 

Florence, Italy 
Anthropologic, Ecole d% 15 Rue Ecole de Medicin, Paris 
Australasian Association for the Advancement of Science, 5 Elizabeth Street, 

Sydney 
American Oriental Society, 245, Bishop Street, Newhaven, Conn., U.S.A. 
American Philosophical Society, Philadelphia, U.S.A. 
American Museum of Natural History, Washington 
Asiatic Society of Bengal, 1, Park Street, Calcutta 
Anthropological Department, University of The Philippines, Manilla. 

Bataviaasch G^nootschap, Batavia, Java 

Bureau of American Ethnology, Smithsonian Institute, Washington 

Bemice Pauahi Bishop Museum, Honolulu, H. I. 

Canadian Institute, Ottawa, Canada 

Canadian Department of Mines, Ottawa, Canada 

Dominion Museum, Welling^n 

Ethnological Survey, Manilla, Philippine Islands 

Fijian Society, The, Suva, Fiji Islands 

General Assembly Library, Wellington (two copies) 

Geographic, Soci^te de, de Paris, Boulevard St. Germain, 184, Paris 

Geographical Society, The American, Broadway, at 156th Street, New York 

High Commissioner of New Zealand, 13, Victoria Street, Westminster, London 
Historical Society, Honolulu, Hawaiian Islands 

Institute, The New Zealand, Wellington 

Japan Society, 20, HanoVor Square, London, W. 

Kongl, Vitterhets Historic, och Antiqvitets, Akademen, Stockholm, Sweden 
Koninklijk Instituut, 14, Van Galenstratt, The Hague, Holland 

Na Mata, Editor, Suva, Fiji 

National Museum Library, Washington, U.S.A. 



Digitized by 



Google 



LIST OF EXCHANGES, xi 

Peabody Museum of Arohseology, Harvard University, Cambridge, U.S.A. 
Philippines, Bureau of Science, Manilla 

Queensland Museum, Brisbane, Queensland 

Royal Geographical Society, Kensington GK>re, London, S. W. 

Royal Gkogpraphical Society of Australasia, Brisbane 

Royal Geographical Society of Australasia, c/o G. Collingridge, Waronga, 

N.S.W. 
Royal Geographical Society of Australasia, 70 Queen Street, Melbourne 
Royal Society, Burlington House, London 
Royal Anthropological Institute of Great Britain, The, 50, Great Russell Street, 

London, W.O. 
Royal Society of New South Wales, 6, Elizabeth Street, Sydney 
Royal Colonial Institute, The, Northumberland Avenue, London 
Royal Geographical Society of Australasia, Adelaide 

Smithsonian Institute, Washington 

Societe Neuchateloise de Geographic, Switzerland 

Societe d' Etudes Oceanienne, Tahiti Island 

Tokyo Imperial University, Tokyo, Japan 

University of California, Library Exchange Dapartment, Berkeley, California 
University Museum, 33d and Spence Streets, Pennsylvania, U.S.A. 



Digitized by 



Google 



BOOKS, Etc., RECEIVED IN EXCHANGE, 

GIFT OR OTHEEWISE, IN 1918. 



From Java — 

Bataviaaiioh GenootiMshap. 
From England — 

Royal Colonial Institute 

John Rjlandt Library, Manchester 

Royal Geographical Society 

Royal Anthropological Institute 
From The Philippine Islands — 

The Philippine Bureau of Science 
From America — 

Smithsonian Institution 

American Gteogpraphical Society 

American Philosophical Society 

United States National Museum 

American Oriental Society 

University of Pennsylvania 

University of California 

American Bureau of Anthropology 

Carnegie Institution, Washington 
From Switzerland — 

Sooi6t6 Neuchateloise de G^graphie 
From Fiji— 

The Editor, * Na Mata ' 

The Fijian Society 
From Australia — 

The Australian Museiun 

The Commonwealth Government 

Royal Geographical Society, South Australia branch 

Public Library, Victoria - 
From Hawaiian Islands — 

Pauahi Bishop's Museum 

Hawaiian Agricultural Experiment Station 
From France- 
Society d' Anthropology de Paris 

Society de G^graphie 

Ecole d'Anthropologfie de Paris 
From Holland — 

Koninklijk Instituut 
From New Zealand — 

The New Zealand Institute 
From Italy— 

Societa Italiana D'Anthropologia 
From Tahiti- 
Society d'Etudes Oceaniainne de Tahiti 
From Ed. Tregear, The Rev. P^re H. Audran, Elsdon'Best, W. W. Smith, Capt. 

W. Waller, W. Churchill, W. M. Borge, B. Glanvill Comey, and S. 

Percy Smith 
From W. L. Newman and S. Percy Smith — Admiralty and other Charts. 



Digitized by 



Google 



THE LAND OF TARA 
AND THEY WHO SETTLED IT. 



THE STORY OF THE OCCUPATION OF TB WHANGA-NUI-A-TARA 

(the GREAT HARBOUR OF TARA) OR PORT NICHOLSON, 

BY THE MAORI. 



By Elsdon Best. 



PART VI. 

(Continued from page 177, Vol, XXVII.J 

MAORI OCCUPATION OF WELLINGTON DISTRICT. 
NOTES ON SOME ARCHiEOLOGICAL REMAINS, CONTRIBUTED BY 

H. N. McLeod. 

Oron^^orons^o. Eemaius of native occupation in former times are 
to be seen at this place, and a number of stone adzes have been found 
here. One seen by the writer was nine incites long and three wide. 

Pencarrow Head. Indistinct traces of native occupation have 
been noted on the hill near the lighthouse. 

Point Arthur. Signs of old time occupation seen here. 

Rona Bay. In this and adjacent bays a number of stone imple- 
ments have been found. 

Day's Bay. On the ridge to the north may be seen the remains 
of a fortified position, as evidenced by levelled hut sites, an earthwork 
defence, and butts of tatara posts. 

York Bay. iSigus of native occupation on hill and beach have 
been here seen. 

Lowry Bay. The headlands or hills to north and south show 
signs of occupation, while old ovens have been seen on the fiat near 
the beach. A shell midden of the talus type is seen by the side of the 
road as one proceeds to Waiwhetu. 

N^a Uransra. Teains ago signs of occupation were seen on the 
hill whereon the fort is situated, a])ove the former position of the 
Wharepouri cenotaph. 

Kaiwharawhara. Signs of occupation were seen here in past 
years, apart from the Ngati-Awa hamlet occupied when the first 
European settlers arrived. 'J'hese last native dwellers here had potato 
gardens on the range above the village. 



Digitized by 



Google 



2 JOURNAL OF THE POLYNESIAN SOCIETY. 

Thorndon. We have already seen that a number of small 
Ngati-Awa hamlets were settled along bluff and beach from Pa-kuao 
to Kumutoto, but apparently this part was not much occupied in pre- 
Ngati-Awft days. A few signs of occupation were seen by early 
settlers, but some of tliese might be the result of Ngati-Awa 
occupation. 

Point Jernin^ham. Omaru-kai-kuru. The tokens of former 
occupation seen here were hut sites and a shell midden. Of the former 
some traces still remain. 

Evans Bay shores. On a fair slope above the steep bluff 
about ten chains north of the U.8.S. Coy. Laundry are some fairly 
distinct narrow terraced formations that are probably old hut sites. 

On the ridge extending downward from Mt. Victoria to Te 
Akau-tangi, the bluff above the lower end of Wellington lload, signs 
of occupation were visible in several places. One of these was the 
knoll above Arawa lload, another east of Rata Road ; others on the 
spurs on each side of the gully near Kilbirnie Reserve. Tlie bluff 
immediately overlooking the Reserve showed until lately a number of 
small artificial terrace formations, hut sites of the men of yore. In 
the sixties could be seen signs of old time occupation above high-water 
mark on the shores of the little bay, now reclaimed, extending from 
the tram waiting shed southward. Hut sites also appear to be trace- 
able on Te Ranga-a-Hiwi, at Aka-tarewa, and on the spur running 
down toward the hospital. 

On an old plan of Port Nicholson district, issued by the N.Z. 
Company, the words "coal has been found here " adorns the fore- 
sliore at Wellington Road. That coal has not been rediscovered yet. 

The small stream from Moxham Avenue that runs into Evans Bay 
near Wellington Road is marked Good Water on Captain Herd's 
chart of 1826. A stone adze was found here. 

Miramar Bay. The little cove just below the old Crawford 
homestead. Here, on the slope above high- water mark, as also on the 
northern extremity of Rabbit Hill, which thrusts Shag Point out into 
Evans Bay, are, or were, shell middens, ovens and hut sites showing 
that this was a favoured place of residence in olden days. In the 
hollow on Bridge Street was the Miramar Lagoon, on which a 
pleasure boat was kept for some years. In its former bed were found 
two stone adzes, and also human remains. Excavations along Bridge 
Street have clearly shown that, at some time in the past. Rabbit Hill, 
over which Tirangi Road passes, was an island. 

RonsfOtai. The ridge extending from the block cutting near 
Miramar wharf southward. At several places along this ridge shell 
middens and levelled hut sites tell of former occupation (see map). At 
the southernmost knoll ovens and human remains were found in 
addition to shell middens and hut sites. The next knoll northward 



Digitized by 



Google 



The Land of Tara. 3 

also allows signs of occupation, while between the two knolls is a 
singularly level area, which may have been occupied, or may have 
served as a plaza, or a cultivation ground. Tlie site of the old Crawford 
homestead is also said to have shown signs of former occupation. 

Matlptlia. This old stockaded village, alread3' mentioned, 
covered about two acres at and near the deep cutting through the 
above ridge. About a dozen butts of the old totara posts of the 
stockade have been found in late years; in the forties they were 
plainly visible. One dug up in 1906 was 4^^ feet in girth, and is now 
in the Newtown Museum. Two stone adzes were found here, one of 
which was of greenstone ; also a grinding stone, with human and other 
remains. Food storage pits were also in evidence. On the border of 
Bumham Water, below Maupuia, the skull of a moa was found. 

The Rongotai ridge was occupied long generations ago by the 
Ngati-Hinepare clan. The famous Nepia Pohuhu, of Wairarapa, a 
man learned in the ancient lore of his people, was a member of this 
clan. Further along this ridge, toward Mt. Crawford, several pleices 
show signs of native occupation. Shell middens were formerly visible 
at Shelly Bay, and at several other places at the base of the ridge. 

Btirnham Water. Rotokura or Para Lagoon. Drained by the 
late J. C. Crawford in 1847-49 by means of a tunnel piercing the 
ridge, thus allowing the lagoon waters to flow into Evans Bay. On 
the northern border of this lagoon a greenstone adze was found, and 
other stone adzes at Rima Street and Ira Street 'East, also the tusk of 
some sea creature at Park Eoad. At the junction of Devonshire 
Road and Princes Street a rib bone 15 inches long was found 20 feet 
below the surface. Evidences of old occupation were noted at George 
Street and May Street, and some other places ; also on the hill top at 
Old Farm Road and Kings Road, and the cave at its base. Our early 
settlers found patches of bush in the gullies at the northern end of the 
lagoon, and those clumps of bush were occasionally frequented by 
pigeons and kdkd. 

Point Halswell. A number of hut sites, represented by small 
terraced formations, have been located on the spur extending upward 
from this point. Indistinct remains are, or were, seen at what is 
tJiought to be the site of the old time stockaded village of Mataki- 
kaipoinga. A few cliains westward of this place, on a jutting spur 
north of Shelly Bay are a number of hut sites ; old ovens are also in 
evidence. 

Owing perhaps to the rocky nature of the ground we see nowhere 
in this district any considerable terrace formations such as are seen in 
many other places. No long continuous teiTaces are here seen, but 
merely linchets of small area, often only large enough to accommodate 
one or two huts ; occasionally one may be seen fifty feet in length ; 
few are longer. 



Digitized by 



Google 



4 JOURNAL OF THE POLYNESiMN SOCIETY. 

Kau-whakaarawaru is said to have been a village in Kau Bay, 
immediately east of Point Halswell. Shell middens show that the 
place has been occupied. 

Te Mahan^a. Talus middens are in evidence here, shell and 
oven refuse. Small terraced hut sites were at one time visible on the 
hills above, where, in excavating operations for the modern fort, the 
butts of some totara posts are said to have been unearthed. In the 
waters below a tanitoha or water monster is said to have abode in days 
of yore. 

Between Te Mahanga and Te Karaka or Karaka Bay is the place 
we call Searching Bay, where many signs of native occupation have 
been seen, including such refuse of btmes of man, fish and birds, 
including nwa bones, together with shells and oven stones. On the 
slopes above .ten'aced sites were noted^ as also on the crest of the 
ridge between Crawford and Fortification Eoads. Hard by a grinding 
stone was found. 

Karaka Bay or Te Karaka. Here over a considerable area are 
signs of occupation ; evidently this was a favoured place of residence. 
Large quantities of shell and oven stones on and below the surface, 
and a number of implements of the neolithic Maori have been 
recovered here, such as stone adzes and chisels, bone combs and 
tattooing implements. Here also was found a fine piece of greenstone 
SJlbs. in weight, in a partially ground condition; apparently it was 
intended to fashion a -^nere therefrom. 

At the entrance to a cave near the wharf were found human bones 
and a stone chisel. At the point just south of Ktiraka Bay and north 
of Worser Bay, which Rangiwhaia Te Puni maintained is the true 
location of the name Taipakupaku, terraced hut sites are seen on the 
ridge. Here also a human skeleton was disinterred, by the side of 
which a stone adze was found. See "Transactions of the New 
Zealand Institute," Vol. XXXII., p. 271. Other human remains 
have been found here, as also some stone adzes, bone needles and 
other objects. South of the point eight human skeletons were 
unearthed during some excavation work. 

Kakariki in Worser Bay. The small flat-topped spur point 
here has been a fortified hamlet in the past, as shown by levelled top, 
terraced sides and fosse whereby it was cut off from the ridge. Hard 
by were found two human skeletons, also a very fine stone weapon 
{patu) of the ^nere type, which is preserved in the Board Room of the 
Wellington Harbour Board. On the north side many scattered and 
broken human bones have been found. 

The famed fortified village of Te Whetu-kairangi is said to have 
been situated about the site of the State School on the ridge above the 
Bay. At the base of the bluff, near the spring Te Puna a Tara, or 
Te Puno a Timirau, have been seen signs of old-time occupation. 



Digitized by 



Google 



The Land of Tara. 5 

Across Seatoun Heights Koad from the School is a knoll which 
carries marks of human occupation, as also does the spur soutli of it. 

Seatoun. On this sandy flat, and near the hill numbers of old 
implements have been found, some of which are in tlie local Ghnstie 
Collection in Newtown Museum. Among the discoveries may be 
cited adzes, stone flake knives, tish-hooks, pendants, needles or 
awls, spear points, bone toggles, a fine moa bone ripi paua^ 
cloak pins, as also a cut human jaw bone, other human hones, and 
moa bones. Several implements were found in a deep rock cleft. 
There has been much native occupation of this flat in past times. 

On the highest part of the ridge above the flat, as also half-way 
up, hut sites have been noted, as also on the spur terminating at Ira 
Street and Broadway junction. Above Church Street are terracings* 
At the quari'y, below the Church Street level, a cave was opened up 
which at some time in the past had been both accessible and occupied. 
Herein was found the biggest collection of moa bones found on the 
peninsula, with bones of other birds and of seals, as also human 
bones. The moa bones are preserved by the Miramar Borough 
Council. 

A number of middens were foimerly in evidence on Seatoun Flat. 
Remains of human skeletons, scattered skulls, etc., have been found 
here, also a fine greenstone ear pendant. Old native ovens were seen 
on the slope near tiie tunnel, anil on the ridge above are signs of 
occupation for some distance in the form of terracings and small 
levelled areas, evidently hut sites. Other such traces are seen on the 
hill west of the signal stations. It is now impossible to tell which 
of these occupied places were open settlements, and which were 
defended by stockades, save that some of the places occupied on fairly 
steep slopes could not have been defensible positions. 

Ortiaiti. Fort Dorset now occupies the site of Oruaiti pa, one of 
the old stockaded villages of past centuries. Prior to being interfered 
with the ridges showed many levelled hut sites, sufficient to accommo- 
date about fifty huts. A number of water-worn boulders scattered 
about were probably used as blocks on which to pound fern root, and 
for other purposes. Butts of posts, one of which is in the Newtown 
Museum, were found here, as also stone flakes and other tokens of 
former occupation. Some information concerning this place, and 
other matters pertaining to Hataitai are contained in the works of 
Jas. Mackay and Canon Stack. At one time Oruaiti is said to have 
been occupied by the Rangitane folk. 

Paewhenua. In the bay beh>w the Signal Station signs of 
occupation have been noted in the form of shell heaps and human 
remains. The tooth of a sperm whale, partly cut through, and half a 
stone mere, bored by marine creatures, were found on the beach. At 
one time a considerable number of karaka trees grew along this coast, 



Digitized by 



Google 



6 JOURMML OF THE POLYNESIMN SOCIETY, 

but many have disappeared and otliers are dying, as at Te Bimnrapa. 
The finest tree of this species in the district in the sixties was that 
which stood at Nga Uranga. 

The raised beaches along this coast line are plainly discernible^ 
and form an interesting study. The pieces of flint found on the beach 
in this vicinity have apparently been brought hither in late times in 
some unknown manner, otherwise flint flakes would assuredly be 
nmch more numerous at places formerly occupied by natives. 

Tarakena. The three points which form the southern portion of 
the peninsula bear marks of former occupation. The termination of the 
eastern spur shows a considerable number of terraced hut sites, much 
weathered. The western spur is but little marked; the middle one 
bears the plainest g^oup of hut sites to be seen in the neighbourhood. 
Butts of stockade posts, shells and oven stones are seen here. In the 
mouth of the gully below, the original Pilot Station is the best 
preserved midden of the district. Here many implements have been 
found by Mr. H. M. Christie, the writer, and other seekers, such as 
stone adzes, chisels, pounders and grinding stones, stone cutters, flake 
knives, bone spear points, also pendants, and carving implements and 
fish hooks, with teeth and bones of dogs, birds and fish. Mr. Green 
here found a fine greenstone chisel, and also a curious carving in 
soapstone, the design being that of a human figure reclining on the 
back of some creature, presumably a whale. Students of Maori 
myths will recognise the meaning of the design. A somewhat 
similar figure is said to liave been found in the South Island. 

The following remarks on the site of the old pa on the hill on the 
western side of this little bay are from one who made a close 
examination of it : — This is the best preserved of the old pas of Miramar. 
The upper part of the position, overlooking the beach, has been 
scarped ; on tlie northern slope a part of the scarp, about five to six 
feet in height, is still extant, also another piece on the eastern slope ; 
the other two sides being precipitous. There are six small residential 
sites, artificial terraces, at the upper part of the position, and seven 
more on the eastern slope to the bluff head, the largest of which is 
about 19 yards long and 3 to 4 yards wide. The remains of the scarp 
on the northern side still shows signs of having had a ditch or small 
fosse at its base. The terraces of the upper part would accommodate 
about twelve small huts. At the western extremity of the pa summit, 
above a small saddle in the spur, the butt of a half decayed totara 
post, 10 inches in diameter, is still iu position. It may be one of the 
original stockade posts. Another near the bluff head is probably 
modern, of the Pilot Station days. It is probable that the folk who 
lived hard by the midden in the mouth of the gulch below, were the 
inhabitants of this /?fl5, which served as a refuge v'heu enemies 
appeared. . 



Digitized by 



Google 



The Land of Tara. 7 

Palmer Head. This point is on the eastern side of tlie guloh of 
Tarakena, and the high bUiff-browed hill shows a number of hut 
sites, now much eroded, but still recognisable. On the eastern side of 
the extremity of the spur is a ditch-like depression that may be an 
old entrance way from the beach below. If this place was ever 
defended by stockades, then the lines of defence would include the 
iirst knoll on the ridge so as to take advantage of the depression or 
saddle just beyond it. Hut sites also seem to be in evidence further 
up the spur. Up the gully from Tarakena runs an old road made in 
early days to give access to the Pilot Station. 

Proceeding along the beach westward of Tarakena we come to the 
stone quaiTy, near which some moa bones were found below the road 
level. 

On the crest of the hill at Hau-te«taka, the eastern headland of 
Lyall Bay, have been found many fragments of moa q^^ shell, as also 
on the sandy area on the lower levels. Old ovens have been seen on 
the summit of the headland, which is pretty sure to have been 
occupied in former times, for it is just such a position as appealed to 
the neolithic Maori. Erosion and dinft sands have, however, con- 
cealed all evidence of such occupation, save the umu or oven. In the 
first, second and third gullies east of this hill a number of interesting 
finds have been made, including shell middens, implements, human, 
whale and tuatara bones, with fragments of moa egg-shell, and 
probably tlie gizzard stones of that creature. Large wliale bones have 
been seen as much as 140 feet above high- water mark oti. the sandy 
slope of the second gully ; others on the sands between the two first 
gullies. Wlien exposed these bones soon crumbled away. Further 
on toward the Golf Links, at the base of Green Spur, the long spur 
trending down from the Orongo Ridge, are tokens of foimer occupation 
in the form of ordinary village refuse, stone and shell. 

Lyall Bay. The name of Hue-te-para assigned by Crawford to 
the foreshore and sandy beach is not recognised by any Ngati-Awa 
or Wairarapa natives who were questioned twenty-five years ago, nor 
is the name Tapu-te-rangi known to them. Some interesting middens 
were formerly in evidence on the isthmus, and one still exists about 
ten chains south of Kongotai Terrace. On the eastern side portions of 
chaiTed moa bones and pieces of egg-shell have been found a few 
chains from the beach. Moa and human bones have been found on 
the sands in past yeai's. Mr. W. Capper has found numerons 
implements, a carved piece of whale's bone, twelve inches long, at the 
foot •f Moa Hill, as the headland hill above Hua-te-taka is sometimes 
called ; also a moa skull and toe bones near the quarry, moa bones and 
shell fragments at Maranui and the east side gullies, as also some 
stone adzes, one of which is greenstone. Most of these objects went 
to England. Mr. Bourke found a piece of carved wood, probably 



Digitized by 



Google 



8 JOURNAL OF THE POLYMESIAM SOCiETY. 

belonging to a canoe on tlie isthmus. Mr. A. Hamilton found a well- 
worked piece of greenstone, mea egg-shell fragments, and jaw bones 
of tuatara near the site of Maranui School. Sand cut stones of 
curiously symmetrica] form have been found in numbers on the 
isthmus, and the raised beaches of this area are an interesting feature 
of the place. Many stone knives of flake foim have been found in 
common greywacke, a few flint specimens, and some obsidian knives. 

Te Ran^a-a-Hiwi. We have seen that this is the Maori name 
of the range extending from Pt. Jerningham to the coast between 
Lyall and Island Bays, and on which are the three peaks known as 
Mt. Victoria, Mt. Alfred and Mt. Albert. Some traces of former 
occupation on this range we have already noted ; a few more remain 
to be mentioned. On the slope north-east of Newtown Park, and on 
a spur above Maranui School, such signs are seen. On Queens Drive, 
south-east of the school, a village lias existed in the past. Along the 
old beach levels at the hill bases in Lyall, Haughton and Island Bays 
have been seen old ovens and other tokens of native occupation. On 
the spur above the Corporation stone quarry, western side of Lyall 
Bay, a number of small tenaced hut sites formerly existed, but many 
have been destroyed. A stockaded village has undoubtedly existed 
on this hill, as shown by the remains of a defensive trench at the 
upper end of the spur, where it juts out from the hillside. It is a 
similar position to the one on the bluff north of Days Bay. 

Haewai or Haui:hton Bay. Here we see that no suitable sites 
were available near the beach, but signs of occupation were formerly 
observed by the streamlet at the head of the bay. On the liills above, 
however, a number of old hut sites are still in evidence. On the 
western slope of the ridge that sepai-ates Lyall Bay from Haughton 
Bay, near the point known as Te Rae-kaihau, are a num})er of small 
terraces on a small spnr offshoot above Haughton Bay. On the steep 
slopes on the western side of the bay similar sites are seen. In nil 
cases these terraced hut sites would be wider wlien occupied than they 
are now, owing to several causes. 

Island Bay. Prior to Euroi>ean settlement traces of native 
occupation were di8cemi>)le all round the bay, on the flat and the 
hills on both sides; it appears to liave been a favoured ]>lace of 
residence. A number of stone implements have been found here, 
some of which are in Mr. Beckett's collection. Old ovens and midden 
refuse of shell, bone and stone, including human bones have also 
been seen in the past. Small terraced hut sites are seen at Urnhau, 
the high hill on the eastern side of the flat, on the hillock above 
Liffey Street, and on tlie central one ; at Milne Terrace and on the 
hill at High Street. The teiTaced knoll above Cliff House may have 
been sin'ix>unded by a stockade. The island also retains impressions 
of man's handiwork, both on the central butte or hillock, and below 



Digitized by 



Google 



The Land of Tara, 9 

it, where the piled up stones possibly surrounded huts with sunken 
floors. On the eastern side of the Tawatawa range, further north, 
a spur jutting out from Vogeltown may also have been occupied. 
Part of a broken p((tu (a short stone weapon) found on tlie island is 
pr<)ba})ly a relic of some old time fight. 

Owhiro. This has been another much favoured place of residence, 
and two middens were in evidence until lately. Here a number of 
stone implements have been found, including the smallest and most 
interesting greenstone gi*aver known. As late as 1915 a number of 
interesting relics have been found here, many stone knives and flakes 
bearing the mark of percussion, worked bones, and an autmu or oclire 
muUer, etc. A village site, with its debris is on the hill on the eastern 
side of the Owhiro road, with its midden below. On the western side 
a talus midden shows that the spur above was occupied. The flat on 
the western side of the road, north of the bridge, has pro})ably been 
cultivated, as food storage pits are, or were, in evidence near the 
creek. 

Sinclair Head. On the eastern side of the point a midden was 
formerly visible, though now obliterated >)y debris from the hiUside. 
On the hill above the point are much weathered hut sites. 

Waikomaru pa. This small position showed, as late as 1911, 
post butts, hut sites and shell refuse. The hamlet must have been a 
very small one. 

Karori Stream. At the mouth of this stream a village has 
stood in the past, and signs of occupation are seen in other places in 
the vicinity. A considerable amount of village debris is still 
observable, and a number of old implements iiave been found here. 
The Opuawe hamlet, far up stream, was occupied in European times, 
as shown by the peach trees growing there many years ago. 

Waiariki. This place was occupied for some time after the 
arrival of Europeans. The small hill by the stream side was possibly 
defended by a stockade in pre-Ngati-Awa times. 

Oteron^o. Here many signs of occupation are still seen on the 
shores of the bay and on the banks of the stream. A number of 
implements have been found here. At other places on the coast small 
middens betoken native occupation. 

Ohatl. The shores of the bay carry signs of much occupation in 
tlie form of middens, ovens, etc. The taldeland of Te Bama-a-paku 
has also been occupied. 

Te Ika a Marti. This place shows the only old fortified position 
in tlie district wliich is entirely surrounded by earthwork defences, 
which consist of rampart and fosse. On the central spur facing the 
bay tliis position measures some 80 yards. The fosse still shows a 
depth of 4^ feet, and the rampart a height of 4 feet, tJiougli erosion 
has played sad havoc. Pits, some post holes, and waterworn boulders 



Digitized by 



Google 



10 JOURMML OF THE POLYMESiMM SOCIETY. 

are the only other tokens of former occupation. On the hill to the 
west of the station homestead other si^iis of native occupation are 
seen. 

Owhariti. Bay. On tlie point at tiie western side of the bay is a 
well preserved pa, name unknown. The earthwork defences of wall 
and ditch across the base of the point were in good preservation 
25 years ago, and are still in fair condition. Butts of stockade posts 
were seen here, and shell refuge is in evidence all round the bay, 
where a number of places have been occupied in past times. Signs of 
occupation have also been noted above the beach south of the bay, 
towards Te Ika-a-Maru. 

With the exception of the little Ngati-Awa hamlet in a clearing at 
Opuawe, it will be seen that all the native settlements were (m or near 
the coast. With the exception of the Miramar Peninsula, the Ran»^u- 
a-Hiwi ridge, and a smaller area at Te Aro, the whole of the main 
peninsula was forest covered in the old Maori days. Most of the 
native settlements were on the outer coast line, either just above high- 
water mark or on the hills overlooking the beach. There was little 
occupation of the western side of the Wellington Harbour, apparently, 
in pre-Ngati-Awa days, but more on the eastern side, and as far as 
Hutt Kiver. This manner of occupation would be owing largely to 
the fact that a great part of the food supplies of the people must have 
been obtained from the ocean. This district can never have supported 
any large population, such as did the fertile lands of Taranaki, 
Turanga, Whakatane, the Auckland isthmus, Taiamai, Oruru, and 
some other areas. It was not a suitable place for the cultivation of 
the sweet potato, which could never have been an important part of 
the food supply. A certain quantity of food stuffs, principally birds, 
would be obtained from the forest, l)Ut the procuring of these supplies 
necessitated nothing more than temporary camps in the bush ; there 
were no permanent hamlets within the forest. The aim of the people 
was to preserve the forest, not to destroy it. 

In the Dominion Museum are preserved some old native imple- 
ments found in this district, including the following objects : — 
Stone adze found in (old) Government House grounds. 
Stcme adze found on site of the museum. 
Stone ndze found at Pipitea. 
Stone weapcm {f>atu onerva) found at Island Bay. 
Stone adze found at Lowry Bay. 
Stone flake implements from Miramar and Owhiro. 
St(me hammers from Owhiro. 

Shell trumpet {Septa ruhictindum) from Somes Island. 
Bone fish-hooks, barbs, shell teeth (pendants). Miramar. 



Digitized by 



Google 



The Land of Tata. 11 

Stone fibre beater from Pauatahanui. 
Stone adzes, etc., from l^itahi. 
In the local collection made by Mr. H. M. Christie, and now in 
tlie Newtown Museum, are some interesting relics of neolithic times, 
including stone adzes, chisels, sinkei^s, and borers or drill points, also 
fiakes of flint, obsidian, &c. ; bone fish-hooks, barbs, awls, needles, 
spear points, cloak pins, &c. Human bones and shell teeth (the latter 
used as pendants) are also included. The piece of obsidian marked 
* tomahawk ' can never have been employed as such a tool, however. 
The rude hand-made iron axes or adzes are a very interesting relic of 
early European trade, as also are the locks of flintlock muskets. 

Mr. Beckett's collection of old atifacts found in this district is 
probably the best ever made in this area. 

« « « « 

In any examination- of the sites of native settlements in the 
vicinity of Wellington, the observer is impressed by two facts, the 
very few signs of hamlets having been fortified, and the situation of 
a number in places that could not [)08sibly have been defended. The 
evidence before us seems to show that the people of this district were 
never so nmch harassed by the raiders as were those of many other 
places. One of the principal causes would be that occupants of this 
area were, in most cases, nearly related to those of the Wairarapa 
district, hence most of their quarrels were with Muaupoko of the 
Otaki district, and other tribes to the north of them. Hamlets 
situated in the mouths of narrow gullies, such as existed at Tarakena, 
or on slopes such as Owhiro, would be indefensible, yet the middens 
at such places call for prolonged occupation. Doubtless the men of 
yore lived much of their time at such places, as they also did at 
Porirua, but, on the approach of enemies, retired to stockaded 
positions, or took refuge in the forest. Presumably stockade defences 
were employed owing to the rocky nature of the land which, in most 
places suitable for hill forts, did not lend itself to the formation of 
ramparts and fosses by means of wooden implements, hence the 
uninteresting aspect of old village sites here in comparison with those 
of other districts. The positions where even a single line of earth- 
work defence was employed, are but few, and consist of one at Days 
Bay, one each at Kakariki, Tarakena, Waitaha and Owhariu, while 
Te Ika-a-Maru is the best specimen. 

It is interesting to note that Cook, who anchored off Palmer Head, 
makes no remark as to seeing any native pa or open village, as he 
seems to have done whenever he saw any. If either the hills at 
Tarakena had then l)een occupied at that time he could not fail to 
have seen the huts and stockades. 



Digitized by 



Google 



12 JOURNAL OF THE POLYMESiAM SOCIETY. 

Tlie story of the settlement of this district, as preserved by native 
historians and given above, is one of much interest. In no other case 
have we gained so clear a description of the settlement of a district, 
and the definite statement made as to Miramar being at that time an 
island much enhance the interest of the narrative. The details of 
construction of stockade defences and of the manipulation of weapons, 
as explained by Whatonga, are the most illuminating notes on those 
subjects ever collected, and the various illustrations <»iven of 
old customs and beliefs throw considerable light on the mode of life 
of the Maori. 

Like all Maori narratives, however, the story has certain 
unsatisfactory aspects and inconsistencies. If we accept all state- 
ments made in the above story we must believe tliat about six 
generations of the Toi family were alive and flourishing at one and 
the same time. This is a common weakness of native traditions. As 
to Whatonga being still alive when the ** Takitimu'^ canoe arrived on 
these shores ; this cannot be accepted. Again we scarcely believe that, 
prior to the deatli of Tara, tlie band of 200 immigrants had so 
increased in numbers as to occupy fire fortified villages, and be able to 
raise so forminable a body of fighting men. This leads to other 
matters of questionable aspect, for the Muaupoko tribe is said in tra- 
dition to have originated a long time after the time of Tara. Perhaps tlie 
most interesting question of all is this one of early inhabitants of these 
southern parts of the island. During the exploration and settlement 
of the district there is no word of any people occupying the more 
southern parts of the island, including the Napier district, j'et our 
party under Tara and Tautoki deem it necessary to construct fortified 
villages, and prepare to defend them. Against whom were these 
local villages so fortified ? Any enemies to be feared in the time of 
Whatonga must have been of the Mouriuri or Maioriori aborigines, 
and there is no evidence to show that these folks occupied any area in 
the southern part of the island. They occupied Taranaki, we are 
told, and the Mamoe clan of that people settled in the Napier district 
apparently after the arrival of Toi o«i these shores. 

There is another point of much interest that may be alluded to 
and that is the fact that we have collected no tradition as to the first 
Maori settlers having seen the moa in this district. Remains of those 
creatures have been found at l)oth Wellington and Porirua, and in 
some cases such discoveries seemed to slipw that the moa had formed 
part of the food supply of residents of this distiict at some time in the 
past. For instance bones and fragments of egg-shell have been found 
near old ovens. Mr. Chapman found moa bones at Paremata said to 
show marks of some cutting instrument. Mr. Beckett found a bone 
near Sinclair Head bearing similar marks. At the same time these 
nicked bones are no actual proof of Maori knowledge of the great 



Digitized by 



Google 



The Land of Tara. 13 

bird in this district. The marks may possibly been made in late 
times, or at least long since the birds disappeared. Certain old 
settlers have remarked that moa bones were seen on hills and in gullies 
west of Owhiro in former years, when the bush was burned off, but 
that they soon disappeared. After the destruction of tlie forest on 
steep slopes the sui*face began to fritter away, and the creeks to 
remove great amounts of debris. The late Mr. Travers found a moa 
leg bone on a hill above the Hutt Eoad, beyond Nga Uranga. Other 
bones have been found at Miramar, Porirua, Pae-kakariki and Wai- 
rai*apa. 

A vast amount of nonsense has been written concerning the hapless 
moff. It has been accused of strolling about Gollan's Valley in 1842, 
and thereby annoying certain veiucious sawyers. See " Titinsactions 
of New Zealand Institute," Vol. XXVI. ; a paper by H. C. Field. A 
still worse case was that in which a stalwart moa, 16 feet liigh, so far 
forgot its proper place as to perambulate tlie Kangitikei district in 
1870. When a journal of this nature publishes such childish fables^ 
it is time for modest folk to retire. 

We are told in native tradition tliat, when the party of Toi 
readied tlie Bay of Plenty, the moa was seen inland of Maketu, and, 
after much trouble, one was trapped by Bua-kapanga, hence the 
saying, " Te Mann nni a Rua-kapanga'*'* (The great bird of Bua- 
kapanga.) Thus we might well expect to hear that others were seen 
in this district when Whatonga and his sons arrived here, more 
especially as we are led to believe that this district was uninhabited at 
that time. But local tradition, so far as it has been preserved, is 
silent as to the lost bird. The old men who have passed away may 
have known something about it, and we must remember that the 
traditions that have been recorded are but a very small part of what 
was known when Europeans first arrived here. 

EARLY VESSELS TO PORT NICHOLSON. 

Native tradition speaks of two vessels as having entered this 
harbour early ia l^st century, of which we appear to have no record. 
In 1878 Karauria Pahura stated that, prior to the time when 
Ngati-Ira were ejected from these lands, a whale ship entered the 
harbour and anchored off Te Korokoro, where a native village existed 
in tliose days, the principal house of which was called Te Putawaro- 
rangi. The vessel lay there for at least several days and took in water 
and fuel, and the captain gave the natives two spotted pigs. He 
cohabited with a native woman named Baumata-nui, whom he wished 
to take away with him, but to this the people objected. This 
incident is mentioned in song. The vessel had come from a place 
called Tiakitene (Jackson), hence a child born at Te Korokoro about 
that time was named Tiakitene. " In after days we learned that 
Poihakene (Port Jackson) was the proper name of that place." 



Digitized by 



Google 



14 JOURKAL OF THE POLYKESIAK SOCIETY, 

Native tradition also mentions a ship, commanded by one Rongotute, 
that is said to have been wrecked at Palliser Bay prior to the Nga- 
Puhi incursion described above, i.e., prior to 1820. 

Te Whaiti, of Wliat&rangi, in Palliser Bay, states that a vessel 
was wrecked in tliat Bay early in the last century, and tliat he has a 
bell obtained from tlie wreck. Tliis bell carries an inscription whicli, 
he has been told, is in French. 

Local natives of Ngati-Awa told the writer many years ago that 
Amuketi, tlie name by whicli Capt. Kent, an early coastal trader, was 
known to tlie Maori, entered Port Nicholson in eary days, a consider- 
able time before European settlers arrived here. We have already 
noted a place at Seatoun named after him. This seafarer entered 
Hokinnga in 1820, was wrecked near Ruapuke, Foveaux Strait, in 
1824, and took Earle and Shand to Hokianga in 1827. 

Local natives also remembered the visit of Capt. Herd to this port 
in 1826. This is the first visit of Europeans to this harbour that has 
been fixed beyond doubt, as will be seen by a reference to the late 
Mr. McNab's " Murihiku," 1907 edition, pp. 364-376. Herd's chart 
of the harbour is an excellent one, and shows about 130 soundings. 
Bumham Water, the small lagoon near the old Crawford homestead, 
and Otari peak are shown on it. He it was who named the port 
Nicholson's Harbour or Port Nicholson. 

When, dunng his third voyage, Captain Cook left Queen Charlotte 
Sound in 1777, he took with him two natives, one of whom he calls 
Taweiharooa. (?Te Waiharaa.) This native told him that a ship 
''* had put into a port on the north west coast of Teei*awitte, but a very 
few years before I arrived in the Sound in the Endeavour, which the 
New Zealanders distinguish by calling it Tupia's ship. At first I 
thought he might have been juistaken as to the time and place ; and 
that the ship in question might be either Monsieur Surville's, who is 
said to have touched upon the north-east coast of Ealieinomauwe the 
same year I was there in the Endeavour ; or else Monsieur Marion 
du Fresne's, who was in the Bay of Islands, on the same coast, a few 
years after. But he assured us that he was not miiitaken, either as to 
the time or as to the place of this ship's arrival, and that it was well 
known to everybody about Queen Charlotte's Sound and TeerawitCe. 
He said that the Captain of her, during his stay here, cohabited with 
a woman of the countiy, and that she had a son by him still living, 
and about the age of Kokoa, who, though not born then, seemed to be 
equally well acquainted with the story 

I regretted much that we did not hear of this ship while we were 
in the Sound, as, by means of Omai (Cook's Tahitian interpreter), we 
might have had fall and correct information about her fi-om eye 
witnesses. For Taweiharooa's account was only from what he had 
been told, and therefore liable to niany mistakes. I have not tlie least 



Digitized by 



Google 



The Land of Tara, 15 

doubt, however, that his testiiuoiiy may so far be depended upon as to 
induce us to believe that a ship really had been at Teerawitte pnor t^ 
my arrival in tlie Endeavour, as it corresponds with what I had 
formerly heard. For in the latter end of 1773, tlie second time I 
visited New Zealand, during my late voyage, when we were continually 
making enquiries about the Adventure, after our separation, some of 
the natives informed us of a ship having been in a port on tlie coast of 
Teerawitte. But, at the time, we thought we must have misunder- 
stood tliem, and took no notice of the intelligence. . . . Taweiliarooa 
told us their country was indebted to her people for the present of an 
animal, which they left behind them, but as he had not seen it himself 
no sort of judgment could be formed, from liis description, of what 
kind it was." 

If this native story contained any truth, the pre-Cook ship must 
have laid in Port Nicholson, Porirua, or under Kapiti ; she would find 
shelter at no other place. The N.W. coast of Tarawhiti would seem to 
imply one of the latter places ! 

Years ago the remains of an old wreck were uncovered at LyalFs 
Bay. In the " Proceedings of tlie New Zealand Institute " for 1892 
occurs the following paragraph pertaining to tlie meeting of October 
26th : — ** The chairman drew attention to some pieces of pottery and 
copper nails, etc., found by Mr. Capper at Lyall's Bay, near the wreck. 
The pottery was carbonaceous, and it was generally thought that the 
nails were of French make." 

Subsequent to the visit of Capt. Herd in 1826, the next vessel to 
enter the harbour, whose visit we are sure of, was the schooner 
" Joseph Weller " of Sydney, in 1835. 

« % * * 

THE COMING OF THE WHITE MAN. 

When European settlers arrived here in 1839-40^ they were well 
received by the natives, and very little trouble arose between the two 
peoples, when we consider the many causes for friction that inevitably 
arise when two races so dissimilar in customs, beliefs, prejudices and 
modes of thought are commingled. The Rangatahi clan, that gave 
some trouble at the Hutt, was not of Ngati-Awa, tliough related to 
Ngati-Tama. 

The cause of this attitude on the part of the local natives lay in 
their position at the time. Many of their fighting men were away at the 
Chathanis, their Kahungunu neighbours eastward were hostile, and 
Ngati-Toa of Porirua were but doubtful friends ; hence the Children of 
Awa were situated between the devil and the deep sea. That is why 
they welcomed the advent of an alien people, who not only proposed to 
settle here, but also brought with them many highly desirable pro- 
ducts — muskets, tomahawks, blankets, etc., not to speak of jews harps^ 
sealing wax and red nightcaps ! 



Digitized by 



Google 



16 JOURKAL OF THE POLYKESIAK SOCIETY. 

When Te Baiigi-haeata sought allies among the Kahungunu clans 
to assist in sacking Wellington and wiping out \\\e pakehaj the answer 
he received Was : — " If tlie white men are expelled, where am I to 
obtain red blankets ? *' {Kei hea he tahnrangi mohu.) 
• In a speech made by Tamati Waka Neiie at the Kohimarama 
Conference in 1860, he said to the assembled natives : — ^^ E nga tangata 

Whanganuiy o Wairarapa^ o Poneke, o Ahuriri! Jcia atatohat koutou ki te 
pakeha. Ki te kino koutou, tnaku e ki atu ki a koutou — e kore taku wahine 
e matau ki te lohatu kakahu, koia ahau i mea ai ma te pakeha e tohatu he 
kakahu mokuV (O people of Whanganui,of Wairampa, of Wellington, 
of Napier, be kind to the Europeans. Should you treat them badly, let 
me say to you — my wife does not know how to weave garments, hence 

1 have decided that Europeans shall weave garments for me.) 



And now we must say farewell to the Land of Tara and they who 
settled it. The bones of the old pioneers of these lands have long since 
mouldered into dnst, the few descendants of the old neolithic ocean 
rovers dwell in the vale of Wairarapa, beyond the rugged ranges of 
the red sun. That sun shines as of yore upon the restless waters of 
the Great Harbour of Tara, but not upon the homes of the Sons of 
Tara, the offspring of Toi, who dared the pathless wastes of Hine- 
moana. For their picturesque stockaded villages have passed away for 
ever, the fair green forests they loved have been torn from the breast 
of the old Earth Mother, whilst the riven and tortured land supports 
an alien and intrusive folk. The descendants of Tara, the explorer, 
and of Ira, the Heart Eater, are unknown in the land, their culture of 
the stone adze has passed away for all time, nought remains of long 
centuries of neolithic occupation save grass grown village sites and 
middens, a few inide implements and place names. 

In the days that lie behind, the Maori had traversed the great water- 
ways of the Pacific, and made known many lands and far scattered 
islets. He explored the rolling realm of Tahora-nui-atea, the far 
spread plaza of Hine-moana, the playground of the Wind Children, 
who come forth from the four quarters to gambol in the vast open 
spaces of the Ocean Maid. In his primitive outrigger he had sighted 
all the isles of the sunlit sea, and made his landfall imder alien stars ; 
he had tied far spread lands together with the wake of his carvel-built 
craft, and carried his ancient tongue from Hawaii to Aotearoa, from 
Rapanui to the Carolines. He traversed his rolling water-ways with 
fine skill and sublime confidence to reach these lands of the far south, 
and spent long centuries in settling them. He brought with him his 
rude neolithic arts as the first wave of human culture, and practised 
them after the manner of his kind and according to his lights. As to 
what plane he may have attained we know not, for the advent of 



Digitized by 



Google 



The Land of Tata. 17 

Europeans called a halt in his progress and shattered the stone age 
fabiic of many centuries. The gulf that yawned before him, traversed 
by us in doubt and ignorance and much suffenng for countless yeai*s, 
was too wide for him to cross in the span of a man's life. 'The fleeting 
years leave him of the stone adze stranded on the shores of the river 
of progi'ess ; across tlie hurrying watei*s we greet the last camp fires 
of the Maori pioneers. 

* * * * 

Of all the scenes familiar to the men of yore in the Land of Tara 
nought remains unchanged save the contour of the great hills and the 
rippling waters of the Great Harbour of Tara. No more are seen the 
hamlets that girt the Red Lake round, the cultivations that fringed the 
Awa a Taia, the paddling of many canoes to the fishing grounds. No 
longer are the fortresses of Motu*kairangi crowded with fierce fighting 
men as of old, ready, at the sign of the signal fires on the Ranga»a- 
Hiwi, to grasp spear and club in defence of their homes. Never again 
shall the chieftain's war canoe swing across the waters of Tawhiti-nui, 
and never more shall the hills of the Land of Tara re-echo the roaring 
choinis of the war song. 

And the Children of Awa, where are they ? Of a verity are their 
numbei*s few in the land. Of all those stalwart, war*seasoned migrants 
who welcomed our fathei*s, none are left. Anon you may see a brown 
skinned descendant on your sti*eets, a lone figure from the age of the 
neolithic, a descendant of the sea kings who ranged a great ocean when 
our forbears were hugging coastlines, a lone figure gazed at curiously 
by passers by. He is not of us, nor of our time ; in the words of a 
survivor of the days of the levelled spear — " Me te mea he xoairua tangata 
4 haere ana '^ — Like a human spirit moving abix>ad ! 

Wherefore have we rescued from oblivian these few fragments of 
the long past history of the Land of Tara, even that the few survivors 
from that past may say : — 

" And from their scholars let us learn 
Our own forgotten lore." 



Digitized by 



Google 



THE FATHERLAND OF THE POLYNESIANS. 

ARYAN* AND POLYNESIAN POINTS OF CONTACT. 

No. 4. 



By S. Percy Smith. 



AS time passes, more and more notes accumulate on the subject of 
jl\. " Aiyan and Polynesian points of contact," and in wliat follows 
some further information on the subject is supplied in continuation of 
papers printed in the " Journal of the Polynesian Society," Vols. XIX., 
XX. It is believed by the writer that further research into this 
question by one who has sufficient enthusiasm, and above all an 
extensive knowledge of Polynesian myths, traditions, legends, folk- 
lore, and customs, and the same of the Aryan Hindus, would lead to very 
great results. But it means years of study, and access to the Sanskrit 
literature, which, it is believed, is as yet not obtainable in this country. 
Dr. Newman has made a start in this direction by the publication of 
his " Who are the Maoris? "• in which he has collected much useful 
information relating to India in connection with Polynesian matters. 
But our old friend will, it is hoped, excuse us when we say he has 
made some mistakes due to the want of a more complete knowledge of 
Polynesian traditions, etc. 

It is probable that in what follows some of the apparent identifica- 
tion mentioned may appear to the reader to be fanciful ; but they are 
the result of an honest attempt to place the source of these old Maori 
traditions in their true bearing. Mr. Ed. Tregear in his many papers 
on the language, and some other points, ought to be consulted in this 
connection. 



H. T. Pio's MSS., Vol. XII., p. 44 (for which the Society is 
indebted to Mr. Elsdon Best), referring to the peaceful nature of the 
people of New Zealand prior to the advent of Toroa and his pai'ty in 
the "Matatua" canoe, in the fourteenth century, says, **That is the 
descent from Toi-te-huatahi (which he quoted) ; it is the descent of 
the Ngati-Awa tribe, and the original people of Aotearoa, or New 

• The word Aryan is here used strictly for the people that invaded India from 
1.500 to 1000 B.C. ; the ancestors of the Hindus. 



Digitized by 



Google 



Aryan and Polynesian Points of Contact 19 

Zealand (i.e., Toi's descendants), they owned Aotearoa, nor did they 
practise evil, and hence the saying of Tuoi when the canoes (of the 
fleet of the fourteentli century) arrived in Aotearoa, and the crews 
commenced to kill Ngati-Tuoi; this was his word, ^ Ahaha! riri 
noa ! Ahaha ! paiu noa ! He aha te take o tenei niahie mahi mat net nga 
iangata o Hawaiki? Kore rawa e mohiotia ana e nga tangata o 
Aqtearpa,^ (Aha ! Anger and killing without reason ! What is the 
cause of the deeds of these men of Hawaiki ? The people of New 
Zealand do not at all understand such proceedings.) The origin of the 
evil doings is from the people of Mataora, of Hawaiki-nui. All evils 
and all good oiiginated in the times of (the gods) Tane, Tangaroa, Tu, 
Tawhiri-matea and Haumia. The evils commenced at the whainga 
(or consecration) of their house (or temple) named Te Tatau-o-Rangiriri, 
and the place where they lived and where the house stood was 
Au-roroa. It was here that everything in the world originated. The 
chief cause of the evil was the destruction of the * vital essence ' 
{patunga i te hau) of the above gods; that was the cause of all evil in 
the world. Let me explain the hau of Tane and the others ; it was 
their mana (power, authority, prestage).* Tane's enemy was Tangaroa, 
and Tu was the enemy of both Tane, Tangaroa and Tawhiri-matea, 
and hence these evils came into the world. Those who escaped (or 
were not subjects of the above evils, or perhaps survived them) were 
the * Heketanga-Rangi ' and the * Hapu-oneone ' who continue to 
practise good works in tlie world. It is said (of them) that (the 
arranging of) peace-making, great and small quarrels, differences 
between brethren" (were due to their teaching, or were their 
principles). 

Such was the teaching of old H. T. Pio, who had been taught in 
his childhood in the Maori College, and there are some tilings in it that 
differ from the usual traditional lore of the Maoris. First we have 
three names of the original Fatherland, two of which are (I think) 
found nowhere else. Hawaiki-nui, or Hawaiki- the -great, is a name 
known to all tribes, and from other accounts implies a continent. 
Mataora and Au-roroa are not sucli wide spread names. In one place 
old Pio says, "The first home of tlie people of Aotearoa (New 
Zealand) was Au-roroa, tlien Mataora, then Hawaiki-nui, then Aotea- 
roa." It is a question if any light can be thrown on the geographical 
position of these places through the meanings of the words. 

Au-roroa : While there are quite a number of meanings to au in 
Maori, there is only one that might be used in a topographical sense ; 
and that is as a * current,' in which case the name means * a long 
current.' This is a very unlikely name to be given to a country, and 

* In auotber place Pio says that the hau was the kwa for which the gods 
«trove, and kura meaus '* knowledge " of a sacred character. This agree^s with the 
account in ** Memoirs Polynesian Society," Vol. III. 



Digitized by 



Google 



20 JOURNAL OF THE POLYKESIAN SOCIETY, 

au^roroa does not seem to belong to the class of words usually 
associated with currents, such as aU'kumey au-rona, au.'tvhtroy etc. It 
might be suggested, perhaps, that it is an equivalent of Taheke-roa 
(which means along rapid), but in the sense in which this latter name 
is used, namely as the '' cnn*ent of death " leading on to Raroheuga or 
Hades, it can scarcely be a proper name for a country. One is therefore 
inclined to think tliat the Maon meaning of au must be abandoned for 
that of the Hawaiian given below. 

We will try to follow tlie word au back along one of the known 
lines of migration of the Maori people, i.e., via Rarotonga, Tahiti, and 
Hawaii. In Rarotonga au means '^ the government," i.e., of a people 
and country by a ruling chief. In Tahiti it has the same meaning as 
in Maori, while hau is " the government," " a reign." In Hawaii it 
means "time," "a reign," "one's life," "a season," and (besides 
others) " a territory," usually where food will gi'ow. 

This last seems tlie only meaning likely to be used in a topo- 
grapliical sense. It is suggested, seeing that one of the Maori 
migrations dwelt for a time in Hawaii, that au as a woixi for 
" territory " was in use there (and long before), and has since gone out 
of use with the Maoris during the 600 or more years since the two 
peoples have ceased to have communication with one another. Hence it 
is thought Au-roroa may be translated as the " (veiy) long country." 

Next as to Mataora ; the second stage of migration according to 
H. T. Pio. This word in Maori means " alive," " in health," and in 
Rarotongan means " pleasure," " pleasant," " happiness," and one is 
inclined to translate the word as " land of happiness," and consequently 
of plenty and safety. H. T. Pio himself says that the name was given 
after the wars refen-ed to later on, and that it expresses the feeling of 
safety, peace and plenty, on the cessation of those wars. 

As to Hawaiki-nui, it is well-known that this is the general name 
of the fatherland of all Polynesians, and is identical with Atia and 
Irihia. It has been suggested that the first part of the name " Hawa,'' 
in Hawaiki, is derived from 8indh-hava, a name for the northern parts 
of India, whUst another of the ancient Maori names of the fatherland, 
Irihia, has been suggested as the equivalent of Vrihia, a name for 
India, or of some part of it. 

The question is of interest as to whether these names can be located 
as an indication of the original home, or fatherland, of the Polynesians. 
No doubt the following attempt to do so will not be considei*ed as a 
proof ; but absolute proof is almost impossible, and thei*e are so many 
things that point to India as their fatherland, that any evidence in 
support of tliat theory ought to be acceptable. 

It is now acknowledged that the Polynesians belong to the 
Caucasian family of the human race, as do the Aryan people of India. 
It is known that the mythology of the former (the Polynesians) has 



Digitized by 



Google 



Aryan and Polynesian Points of Contact 21 

many affinities to the mythology of the Aiyans, and they have, or had, 
very many customs in common. If the Polynesians belong to tlie 
Aryan people, they must have sepamted off froni them in very early 
times, before the rigid caste system of the latter people came to 
predominance, which was after their occnjmtion of the whole of the 
Pan jab. and also after the Aryans came into contact with the dark 
Bharata^ people of Dra vidian origin who were the original inhabitants 
of Inditu This contact took place in the country now known as the 
Pan jab, or north-west India, in the early days of the irruption into 
India; and if the Polynesians formed part of that migration, it were 
better to use the name of Proto-Aryau for them, as indicating the 
contact of the first advance of the Aryans into India from theiir father- 
land called Eran, which — it is suggested — is the equivalent of the name 
of a very ancient country known to the Polynesians under the various 
forms of Herangi, Erangi, Holani, and Harani, according to the part 
of Polynesia from whence each name originates, as recorded in their 
traditions. 

The brief history of the Aryan migration into India is as foUowsf: — 
The dates of the migration into India are variously given as from 1500 
to 1000 B.C., when they crossed the mountains by the Kyber and other 
passes into the Panjab, from Eran (or, as it is sometimes called, Iran), 
and gradually spread eastward to the upper watei-s of the Ganges. 
During the occupation of the Panjab tiie tribes — for they appear to have 
had a tribal organisation at that time — came into conflict with the 
Bharatas or Dasas, or original inhabitants, and fierce fighting took 
place. Many battles were fought, ending in a gradual amalgamation, 
to a certain extent, between the two peoples. These original inhabitants 
are described in the various Sanskrit works of the Aryans, as a very 
dark, or black people, and were much despised and abhorred by the fair 
" Heaven born " Aryans. The eastern part of northern India, which 
was occupied by the Aryans after some centuries dating from their first 
arrival in the Panjab, is described as a richer and pleasanter and more 
wooded country than that of their first settlement. This country was 
much coveted by, and was eventually conquered by the Aryans, and in 
the course of many centuries the people spread down the Ganges to its 
mouth, and all over northern India to the Vindhya mountains that 
partially cut the Indian peninsula in two, the south of the mountains 
being to this day occupied by the Dravidian (Bharata) people. It was 
during the occupation of the Panjab that the priestly craft gained 

* It is from this Bharata people that comes the oldest name for India, viz., 
Bharata-vasha. See Hewitts' ** Myth Making Age," p. 281. 

t See ** Vedic India," by Ragusin ; *' The Original Inhabitants of India," by 
Oppert ; **The Myth Making Age," by Hewitt, and other works of the same 
author; "Sanskrit Literature," by Prof. Macdouuell ; *' Brief History of the 
Indian People," by Sir W. W. Hunter. 



Digitized by 



Google 



22 JOURNAL OF THE POLY IH ESI AM SOCIETY, 

gi-eat ascendancy, and then the rigid caste system of India was 
institnted. 

The suggested explanation of H. T. Pio's story is this: Au-roroa 
(the long teiTitory) stands for the western Panjab with its plains and 
great rivers, where tlie gods-^ofEspring of Heaven, Rangi (the Aryan 
Dyaus*) and the Earth, Papa (called Prithivi with the Aiyans, having 
the same meaning as Papa, broad, extended) — were created. It was 
here the " Heketanga-rangi," or " descendants of heaven," of H. T. 
Pio, a term which the Aryans apply to themselves, came into collision 
with the origiiial people, the Bharatas or Dasas, which are perhaps 
represented by the " Hapu-oneone" of H. T. Pio, meaning "the 
tribe of the soil," or in other words the onginal inhabitants. The wai*s 
that then took place may be represented in Maori traditions by the 
wars of the gods, known under the general name of "Te Paerangi," 
in which the names of twenty-one battles have been recorded in 
Maori history. (See "Memoirs Polynesian Society", Vol. III., p. 
134.) This epoch may be descriptive of the " Wars of the ten Kings" 
of Aryan histoiy. Although the Maori traditions say this series of 
wars was between the gods, and that " they fought as gods," it is 
perhaps easier for people of the 20th century to consider this strife as 
between human beings, and that the record of it has become glorified 
in process of time, and in the form of myth to represent gods instead of 
men. 

In the series of battles " of the gods " — **Te Paerangi" — referred 
to above, the principal enemy of the side that eventually conquered 
(led by Tane, often said to be a god of light) was Whiro, who after 
his defeat became the chief god of Hades. Whiro is often called 
Wliiro-te-tupua, Whiro-the-demon, Whiro-the-uncanny, Whiro-the- 
evil-doer, in which ittpua has many other meanings, as "strange," 
^* gifted witli unusual powers," " evil," etc. Whiro has become the 
god of thieves and evil doers ; the dark night of the moon is called 
Whiro also. He is the great rebel of the Polynesian Myths. 

It would seem probable that this name, Whiro-te-tupua, might 
appropriately be applied to the powei's opposing the Aryans in their 
struggle against the original Dasas or Bharata aborigines of the 
Panjab. Ihipua is just such a name as would be (and has been) applied 
to an uncanny dark race. In the language of myth, Wliiro is the 
enemy of the "children of light"; in other words the opponent of the 
immigrant Aryan people. 

As to the Hapu-oneone, H. T. Pio states that both this people and 
the Heke tan gi- rangi were peace lovers, and a genealogy is on record 

* The name Dyaiis, for the Heaven -father, in later times obtained another 
name Varuna (** the all covering heavens '*). Now in the Ngata-Awa dialect of 
Maori, Wa-runa means the '* space above.'* Probably this similarity of names 
is quite accidental. 



Digitized by 



Google 



Mryan and Polynesian Points of Contact, 23 

from tlie former to the present day, numbering 35 generations. This 
latter record is of no very great value, for it probably is based on the 
same footing as so many genealogies with the name of Bangi, the 
sky-father, at the head of them, and which only means that the names 
following the descent acknowledged Rangi, as the progenetor of all 
mankind. But the supposed peaceful character of the Hapu-oneone 
militates against their representing the Bharatas — and there we must 
leave it. 

The next stage in U. T. Pio's migrating movement was Mataora^ 
which we have seen a few pages back is possibly translateable as ^' the 
laud of happiness and plenty," which, it is suggested, may represent 
the richer country to the east of the Panjab, so much coveted by, and 
afterwards conquered by the Aryans, as referred to above. It has been 
suggested that Mataora is represented by the ancient Indian state of 
Mathura, which is on the Jumna river (a principal branch of the 
Gauges), and now called Muttra, and where the Kuru branch of the 
Aryans lived. It is in the country " coveted by " the Aryans. But 
we do not know enough of the legitimate letter changes between 
Sanskrit and Polynesian to say if the one name may represent the 
other. There is perhaps more justification in supposing that Mataora 
as a descriptive name represents the richer lands of the country east of 
the Panjab, watered by the many branches of the upper Ganges; 

It is perhaps possible that the ascendancy of the priesthood among 
the Aryans when caste was introduced, is represented by the 
prominence given to the priests in the Barotongan recitation of the 
classes of people who attended the great meetings for public purposes 
under the rule of Tu-te-rangi-marama in Atia, the fatherland, as 
described in " History and Traditions o£ Barotonga," Part V., which 
ruler was the builder of the temple called " Te Koro-tuatini " ; and, 
we may say, was a king in Atia, the name by which the fatherland is 
known to the Barotongans. This man was afterwards deified as a 
god. 

With regard to the dark or black people the Aryan records speak 
of — the Bharatas or Dasas, or original inhabitants of India — it is 
possible tliey are referred to in the history of the original expulsion 
or migration of the Polynesians from the fatherland, one description 
of which is to be found in "Memoirs Polynesian Society," Vol. IV., 
p. 9. There are ^wq tribes or different kinds of men mentioned, two 
of whom were " lanky, thin people, whilst the three others were a 
black people, one kind was very black : they were not brown like the 
Maoris." It is said above that the Aiyan migration into the Panjab 
abhorred and despised the black aborigines. This disgust at black 
people (Negroes) was quite characteristic of the Maoris sixty to seventy 
years ago. 

It is suggested above that if the Polynesians are a branch of the 



Digitized by 



Google 



U JOURNAL OF THE POLYKEStAK SOCIETY. 

Aryan people, they must have separated from theiu before the rigid 
caste system became so pronounced ; and it is quite clear also that the 
separation took place before Buddhism developed in India (fifth century 
B.C.) for there is no trace of it in Polynesia. Ragusin says, the caste 
system came to pix)minence about tlie end of the Ve<lic age, wliich 
Macdonnell fixes ut about 200 B.C. ; and as the Polynesian pedigi'ees 
leading back to the time of the exodus from (as I suppose) India, and 
from which we calculate dates, fixes that date at 475 B.C. (see 
" Memoirs Polynesian Society," Vol. FV., p. 12) the Polynesians 
knew neitlier of the caste system nor of Buddhism. 

It is clear from the Indian records that the Aryan people gradually 
moved down the course of the Ganges to the sea, and this movement 
according to General Forlong's tables, took place in about 600 B.C. 
He also notes, '^ Time of great disturbances in India 500 to 400 B.C." 
If, as has been suggested above, the Polynesians were the forerunners 
of this migration (or Pro to -Aryans), and they migrated down tlie 
Ganges in 600 B.C., they would not have been affected by the caste 
system, or by Buddliism, which were at first northern Indian 
institutions. 

A tradition of tlie Maoris, told to the writer many years 
ago by the most learne<l man of tlie South Island then living, was to 
the effect that Hawaiki-nui was a tuatchentM, a mainland, not an 
island, that the southern part was mostly plains, with a high ridge of 
mountains to the north, always snow-clad, and throngli whicli country 
ran the river Tohinga, associated with the deluge. This is not a bad 
description in general terms of that part of India, and the river 
Tohinga (wliicli means the Maori form of baptism) is possibly the 
Ganges, a sacred river of tlie Hindus. 

The stoiy about the deluge, however, is an instance of tlie trans* 
ference of an occuiTence localized in another place, of which we have 
such numerous examples. Although the story of the flood is well-known 
and fully descnbed in Aiyan records, it is believed by scholars to have 
been introduced from Mesopotamia, where gi*eat floods in the Euphrates 
and Tigris gave rise to the story, and formed the basis of the Biblical 
account. The Noah of tlie latter account is the Manu of the 
Aryan story, which word in Maori means ''to float," possibly an 
accidental similarity. 

The Proto- Aryans formed, on their occupation of the lower Ganges, 
the people quoted by Logan (tlie Indonesian Ethnologist) as the 
'* Gangetic Uace," from whom he traces the Polynesians and some of 
the most ancient peoples of Indonesia. 



Digitized by 



Google 



Aryan and Polynesian Points of Contact 25 

There is a Maori tradition to the effect that all fish come from a 
spring called Rangiriri,* and Dr. Newman in his bookf says, there i^ 
such a place-name on the Hugli branch of the Ganges above Calcutta, 
There is always some foundation for similar stones, and it is suggested 
that this " spnng " is where the tidal flow ceased on the Hugli, and 
where salt-water fish were first seen by the Polynesians. In Nobin 
Chandra Das's " Ancient Geography of Asia," his map of India shows, 
that in Aryan times, the sea flowed up to where the Mandar hills 
come near the Ganges, or some 200 miles inland of the present coast 
line, forming a great bay now filled up by the delta of the Ganges. 
At the head of this bay he marks a country or district called Vanga. 
Now Whanga is the Maori name for a bay. Is this similarity of name 
purely accidental? and could the Rangiriri "spring" be situated at 
the head of this ancient bay ? It will be^ noticed in the notes from 
H. T. Pio, ante^ he mentions a building named Tatau-o -Rangiriri. 

But there is another and possible explanation of this " spring " 
from which fish originate at Rangiriri. Hewitt in " The Ruling Races 
of Prehistoric Times," p. 220, says, " It was from the belief in the 
life-giving waters as the author of life that the cult of the prophet 
fish god arose. This . . , . was first developed in India where the 
conception was naturally engendered by the annual recurrence of the 
apparent miracle of the birth of fish from the life-giving rain. For it 
is there that water- tanks formed by excavations, or by throwing dams 
across the hollows between hills or rising grounds, are, though dried 
up every year by the heat of the dry season, found to be swarming 
with fish as soon as they are filled by the rains.:|: These fish .... 
proved by actual experiment, have been hibernating during the dry 
season. . . .*' 

The above is just such an occurrence as would give rise to the 
Maori story. And what does Rangiriri mean ? It is the " angry sky," 
descriptive, it is suggested, of the stoims and downpour of rain that 
mark the inception of the monsoon season in India. 

There is one (and one only I think) Maori tradition to the effect 
that a man named Kahukura introduced the knowledge of the 
kumara (sweet potato) to the Maoris. It is said in this tradition that 
he brought it to New Zealand. This, however, is another instance of 
the shifting of locality, as so often occurs. It is suggested that this 

* A place of this name is recorded in the tradition of the Mangareva Islanders, 
but not apparently connected with fish. 

t * « Who are the Maoris ? »' 

\ These south-west winds, that bring the Monsoon rains, are called martu in 
Sanskrit ;. and mdtu is the Samoan for a northerly gale, perhaps derived from the 
Sanskrit, but the direction altered because the Samoans came from the north. 



Digitized by 



Google 



26 JOURNAL OF THE POLYNESIAN SOCIETY. 

Kahukura is the same man who is mentioned on page 12, ^'Memoirs 
Polynesian Society," Vol. IV., and he flourished, according to the 
Rarotonga genealogies (on which Kahukura is also mentioned), in the 
middle of the fifth century B.C., that is, when these people were living 
(as I hold) in the valley of the Ganges, and it was in the same 
generation that the earl}' exodus from there took place to the east, to 
Tawhiti-roa, and other places on the way to the Pacific. Dr. Newman 
(loc. cit., p. 267) mentions that in Bengal the tuber is called kumar^ 
and that it grows wild in Orissa, the country lying to the south-west 
of the mouth of the Ganges. It is suggested that it was the above 
named Kahukura who got the tuber from Onssa,.and introduced it to 
the knowledge of his compatriots when they were living in the Ganges- 
valley. (See also what Dr. Newman says on this subject in the work 
quoted.) 

Just here it will be well to introduce the Maori account of the 
origin of the himara, which is a true myth, but which has underlying 
it in all probability an histoncal basis, couched in the language of 
myth. We quote from the same H. T. Pio's MS., Vol, XII., p. 109. 
He says, " The humara were the offspring ef the star that takes its- 
flight (low down) on the side of the ocean; it is named Whanui 
(alpha Lyrae or Vega). It was his younger brother Hongo-Maui that 
introduced the himara to this world. The " basket " in which he. 
placed those children was his own body. On his arrival he cohabited with 
Tinaku(orPani-tinaku); she was his wife. When she became pregnant^ 
that man said to her, " You must go to the waters of Mona-riki and 
give birth there," at the same time teaching her the appropriate 
karakia. She did so and repeated the karakia as follows : — 

E Pani ! E Paiii-tinaku e ! O Pani ! Pani-of-the-seed- tubers 

Ki te wai opeope ai In the water bring them forth 

Ka heke i tua, ka heke i waho Let them descend behind, outside 

Me kowai P me ko Pani Like whom F like Pani 

Ka heke i takn aro Descended from my front 

Then were born her children named Nehu-tai, Patea, Wai ha,. 
Pio-niatatu, Pou-aro-rangi, Toroa-mahoe, Anu-rangi, and Nehu-tai- 
aka-kura (names of varieties of kumara). Such were the kumara 
offspring, which those ancients appointed for the sustenance of 
their descendants of this world. On their birth Bongo-Maui said^ 
" Now (let us) institute the (ceremonial) ovens, imu'tapu, imu'kirthau^ 
imu'potaka^ imu'maharoa, imii-kohukohu, (He then describes the* usea 
to which these hnm, as he calls them, or wmw, the common name, which 
were for special classes of priests and people at various ceremonies.) 
All of these things originated at Mata-ora in Hawaiki-nui, and when 
Hoake and Taukata came from Tahiti to New Zealand (a well-known 
story) they introduced this knowledge to New Zealand " (somewhere 
about tlie twelfth century). 



Digitized by 



Google 



Aryan and Polynesian Points of Contact 27 

From the foregoing myth we learn that the humara tuber was 
originally the o£Fspi*ing of the star Vega, or Whanui, the positioi) of 
which is about 38° north of the equator, and consequently does not rise 
high in the sky in New Zealand. Fi'om Pio's home in the Bay of 
Plenty it would he seen ^' at the side of the ocean " as he says. It is 
dear from its low elevation that the star myth did not onginate in this 
southern hemisphere. The importance of the stAr Vega is due to the 
fact that in vei*y ancient times this star wa& the whetu o te tau, the 
^* year star," marking the commencement of the new year among the 
northern people, afterwards superaeded by Mataiiki, the Pleiades, 
which gi-oup was used by the. Polynesians to denote the commencement 
of the new year down to the nineteenth centui-y a.d. And, it is 
suggested, it was due to the impoi*tance given to Vega as the "year 
star" denoting the time for prepanng the ground for the kumara crop, 
that it is said to be the parent of the tuber. Who Bongo-Maui was, we 
have nothing to indicate, except to suggest that this may be another 
name for Rongo-marae-roa (Rongo-of- the- wide-spread-courts, or 
fields), the god of agriculture. He married Pani-tinaku, who was the 
real mother of tlie kumara. Now, in the various Sanskrit works of the 
Aryans, we find that Pani was the name given to the oi'iginal 
inhabitants of India, as expressive of their " hard dealings in trade and 
their acquisitiveness," and there is a very pretty stoiy illustrating this 
feature, quoted by Rtigusin (loc. cit., p. 257), taken from the sacred 
books of the Aryans, the Rig-Veda, X., 108, though that story has 
nothing to do with the transfer of agricultural products. If the kumara 
was growing wild, or cultivated, in Orissa, as stated by Dr. Newman, 
in the country of the Bharata aboriginies, it is suggested that the 
Aiyans (and Polynesians) derived the kumara from these people nick- 
named Pani. H. T. Pio says the " birth " (? exchange) took place in 
Mata-ora, which has been suggested as the parts of India lying east of 
the Pan jab. 

Kahukura is associated with the kumara, as will be seen in the 
following part of one of the karaktas, or invocations, used in planting 
this tuber, eveiy thing to do with which was considered as of a sacred 
nature : — 

Tenei te whangai This hi the offering. 

Ka whaugai na That 18 here offered. 

Ko te whangai o wai ? 'Tis the offering for whona ? 

Ko te whangai o Rougo-mai 'Tib the offering of Rongo-mai 

Ko te whangai o wai ? 'Tis the offering for whom ? 

Ko te whangai o Kahukura 'Tis the offering of Kahukura 

Ko te whangai o wai ? 'Tis the offering for whom ? 

Ko te whaugai o Uenuku *TiH the offering of Uenuku 

&c. &c. &c. &c. &c. 

This was the commencement of the ceremonies, when the priest 
offered the marere, or propitiatory tuber to tlie powers above. Bongo- 



Digitized by 



Google 



28 JOURNAL OF THE POLYMESiAU SOCIETY, 

mai is probably the same as Kongo* Maui before mentioned, god of 
agriculture (though Rongo-iuai is usually said to lie a meteor in which 
form that god appears), and Uenuku is another name for the rainbow, 
thus are these two names (Kahukura and Uenuku) and the kumara 
connected with the rainbow in some manner we have not yet got at. 

It is perhaps only natural that a people like the Polynesians (a 
branch, as suggested, of tlie Aryans) should, on first becoming 
acquainted with (to them) a new and valuable food, ascribe its ongin 
to some super-human source, and connect the discovery with the star 
guiding the preparations for planting, and stars are often refeired to as 
deceased ancestors, meaning pi'obably deified ones. The introduction of 
tlie breadfruit to the knowledge of the Polynesians has a somewliat 
similar mythical story connected with it. 

The Turehu, or fair, or white people, mentioned in the traditional 
history relating to the times just preceding the exodus of tlie 
Polynesians from the fatherland — a date fixed by their genealogical 
tables — are difficult to account for. In modern times tlie Maoris have 
come to look on the Turehu, or Patu-pai-a-rehe, as fairies inhabiting 
parts of New Zealand. This localization is chamcteristic of very many 
legends, all the world over; and the fact that the Nine Islanders have 
some of the same stones about the Turehu as the Maoris, proves at 
once that this localization of an ancient legend has taken place. This 
is the description of the Patu-pai-a-rehe as given to the late Sir Geo. 
Grey by the Waikato people : " The fairies are a numerous people, 
merry, cheerful and always singing, like crickets. Their appearance 
is that of human beings, nearly resembling an European ; their hair 
being very fair, and so is their skin. They are very different from 
the Maoris, they do not resemble them." * It is said of them that their 
music, as they played their flutes, was very pleasir.g. Although 
called fairies by Europeans it is obvious the Maori tradition considered 
them as a people, not exactly like themselves, but still human — indeed 
much the same as they considered white people when they first came 
in contact with them. 

The Maori tradition is that their ancestors learnt the art of 
making iishing-nets from the Turehu people. Obviously this indicates 
that tliere was a time when the Polynesian ancestors did not know 
much of salt-water fishing — very naturally so, if our supposition is 
right that they sprung from the inland Aryans ; and the race they 
learnt the art from must have been a sea-faring people. The Maori 

* In a note to be found under the Maori text, Sir G. Grey adds, ** Upon the 
27th October, 1853, Te Wherowhero (head chief of Waikato, afterwards the first 
Maori king) described the fairies as a white race, elegantly clothed in garments 
quite unknown to the natives, and as delighting in music." 



Digitized by 



Google 



Aryan and Fa/ynssmn Faints of Contact 29 

tradition states that a man named Kahukura fiist learut this art from 
the Turehu, and it is suggested that he is identical with tlie Kahu- 
kura who introduced the kumara to the knowledge of his people. K 
BO, this would be some time in the fifth century B.C., when Kahukura 
flourished. 

Kahukura is also a name for the rainbow, and when this appears 
as a double one, the upper bow is said to be a male^ tbe Ipwer a 
female. 

But the question arises, what white, or fair race, this could be ? 
The Aryan records mention more than one fair race, and Mr. Hewitt* 
seems to consider that branch of the Aryans (or perhaps one of the 
northern races, it is not clear which) called Chamar, connected with 
the Yadu Turvasu (who lived on the banks of the Indus) to be an 
early migration of fair people into India. He says (p. 217) ... . 
** coimects them \^ith the very ancient immigrant race of India, the 

beardless Oharmars p. 219 in Chuttisgurh, where I knew 

them best, by their fair skins and th© beauty of their women. '^ 
Another fair people was the Pandyas or Pandavas. The same author 
says of them (p. 40), " The Pandyas or fair {jpandu) men .... Their 
father star, Canopus, controls the tides in Hindu astronomy by 

drinking up the waters of the ocean " which quot-ation also 

illustrates a Maori belief, to the efPect that a monster named Parata 
causes the tides by the inhalation and exhalation of his breath-^ 
identical with the Pandyas' belief. It was these Pandya people who 
held the state of Madura, which (says Sir W. W. Hunter, loc. cit., 
p. 127) was founded in the fourth century b.c. So far as can be made 
put these Pandya people are not Aryans, but rather a northern people 
living among the Dravidians of the south of India, along the coasts, 
Madura being not far from the South extremity of India (Cape 
Comorin). These people apparently were the earliest navigators and 
traders of the Indian seas, obtaining the timbers for their craft from 
the west coast of India, where the forests formerly came down to the 
waters edge. It is tliese Pandyas or Pandavas, the pandu, fair people, 
that possibly are those from whom the Polynesians learnt the art 
of making fisliing nets and, no doubt, the art of sea-faring, which in 
the end they so developed as to carry them all over the Pacific. It 
may possibly turn out that in the word pandu, we have the first part 
of tlie name Patu-pai-a-rehe, a word to which we can otherwise give 
no meaning, though patUy in Niue Island means a chief, with more 
probability derived from whatu^ also meaning a chief. 

There is also another possible, though not perhaps probable, white 
race, that might be that of Maori tradition, that frequented the 
southern shores of India in very ancient times. In ** Journal Royal 

* " The Myth Making Age," p. 215 ff. 



Digitized by 



Google 



30 JOURNAL OF THE POLYMESIAM SOCIETY. 

Anthropological Institute," Vol. XLVIII., p. 176, H. J. Fleure and 
L. Winstanley say, '* . . .we may note that the Milesians (Gaels) are 
said to have visited Taparobane (Oeylon), India, Asia, etc." This 
appears to have been long before Christ, and if these Qaels reached 
Taprobane, a port in the Straits dividing Ceylon from India, they 
might easily have come in contact with our Proto-Aryans. Our 
authors do not indicate how the Gaels reached that part of the world. 

Mention of the monster Parata above, reminds one of the belief of 
an old friend, long since dead, who had deeply studied Maori 
traditions, to the efPect that the '' Waha-o-te-Parata," said to be a 
maelstrom in the ocean, is situated near the mouth of the Persian 
Gulf, where he had noticed the turbidence of the currents in former 
times. It is this '' mouth-of-the-Parata " that is supposed to influence 
the tides, as does Canopus who drinks up the water according to the 
Pandava belief. One naturally wonders whether the Maori name 
Parata is associated with Bharata, the name of the original Dravidian 
people of India, and of the country. 



Such a large number of notes have accumulated on the subject of 
^* Aryan and Polynesian points of contact," that they must be deferred 
to another occasion. 



Digitized by 



Google 



TRADITIONS OF AND NOTES ON THE 
PAUMOTU (OR TUAMOTU) ISLANDS. 



Collected by the Rev, Fire HerviAudran^ of Fakahiva^ 
Faumotu Islands. 

TRANSLATED FROM THE FRENCH BT R. H. ROOKEL, M. A. 



PART IV. 

A GLORIOUS EPISODE m THE LIFE OF MOEAVA. 

[In the following legend P^e Audran furnishes ns with some of the adventures 
of the Paumotu chief Moeaya. It is interesting to note that the Rarotongan 
Traditions hear witness to the existence of this man, and from them we are able 
to deduce an approximate date at which he flourished. First, we may note that 
Moeava's canoe named ** Murihenua/' is known to the Rarotongan's under the 
varient '* Muri-enua,'* which formed part of the fleet of the chief Naea which 
** came from Ayaiki (Hawaii) to Iva (Itfarquesas) and from Iva to Tahiti and 
thence to Rarotonga. This was before the time of Tangfiia and Karika.'' (See 
** Hawaiki," Srd Edition, page 274.) The period of Tangiia and Karika of 
Barotonga has been fixed tentatively as the middle of the 13th century. 

Next we have in the following paper a reference to Hono^ura — or as the 
Rarotongan call him Ouokura — whom, we may infer from the text, lived before 
or at the time of Moeava, and this will ag^ee with the Rarotongan genealogies, 
for Onokura lived five generations before Tangiia, mentioned above. 

In the Rarotongan account of Onokura, which describes his voyage from 
Tahiti by way of Takaau and Te Pukamaru to the Marquesas, we find that at 
the latter island he came across Te Ika-Moeava who is probably the hero of the 
following Paumotu account. Onokura found at Te Pukamaru a Marquesan 
chief named Ngarue with whom some fighting occurred, ending in a peace- 
making, and then the latter accompanied Onokura to the Marquesas. During 
the fighting the high chief Taugfiia-ariki of Tahiti was killed, and in revenge 
Onokura, with his magic sling, sent a stone with such force that Ngarue's canoe 
was smashed. In one of the thirty-nine scmgs that illustrate Onokura*s 
adventures, we find Te Ika -moeava referred to as Moeava, which seems to 
identify the former with the Paumotu hero. The song is a long one, but this is 
the reference referred to r — 

Tangi ua maira i te paii mei tai e — 

Rai ka tari ki o Rongo, ka tari ki o Rongo, 

Nga kai a Moeava, kia katoatoa rei iri 

The drums are sounding from seaward. 
Prepare a great offering to (the god) Rongo, 
Take all the food of Moeava as offerings 



Digitized by 



Google 



32 JOUmiML OF THE POLYHESIMH SOCIETY. 

If we are right in identifying these two individuals, it follows that the 
period of Moeaya is about the year 1125 a.d. — Editob.*] 

A BEMABKABLE COMBAT BETWEEN MOEAYA AND PATIBA 
AT MAKEMO. 

A BOLD and brilliant action, which had Makemo Island for its 
scene, spread the renown of Moeava throughout the whole of 
Eastern Polynesia. The memory of this action is deeply graven in the 
minds of the natives, who still speak of it with pride, and make it 
their boast. 

Moeava, as is well-known, sailing in his **Murihenua,'*^ made 
constant journeys throughout the Tuamotu Group, traversing the 
islands in all directions. One day, while sailing off Kaukura, he 
happened to meet another traveller, who was a stranger to him. 
According to the custom in such circumstances, the two challenged 
each other. Moeava asked the new comer into these parts i ^^ vat to 
ia vaJca i taku tar a neif^^ (What is that boat which is lying thus beside 
me?) 

In a loud and sonorous voice the strainer replied : *^ It is Patira 
whom you see." 

Patira was a famous kaito, a warrior cast in the same mould as 
Moeava. His fame no longer required to be made ; ^s it represented 
him as one of the greatest champions of the isles beyond Tahiti. 
According to tradition, he had sprung from the great host of Marama. 
Q'No roto mat oia i te tint rahi o Marama^) The ancient and famous 
territory of Marama was, it seems, in the Cook Islands or in the 
islands to the leeward, if not still further off. Between Tahiti and 
Sa^atea is *' Te mitt o Marama" (the sea of Marama). In the South 
Island of New Zealand (Te Wai Pounamu) on the east coast between 
Okai'ito and Orepuke is a district bearing the name of Marama. May 
not this tini spring from here ? f Patira, who sprang from it, was a 
valiant warrior of collossal stature. The story goes that, with one of 
his ordinary steps, he strode easily from island to island, aiid that he 
could, without a boat, go the round of all the different islands of our 
archipelagoes, at his own sweet will. 

Prompted by curiosity Moeava proceeded with his questions : 
*^ Where ore you going ? " he asked Patira. The latter replied 
enigmatically, " Te tere ana vau i te kurtkuri o te huraro.^^ (I am 

* Headers must remember that the Paumotu dialect always has the letter 
** n " before the ** g,** thus, ** the tribe of the Goio " must be read ** the tribe of 
the Ngoio.*' — Editoe. 

t We do not think this likely. It is much more probable that Patira came 
from ** the sea of Marama,'' i.e., Ra'iatea, etc., or from the island Whiti -marama, 
flaid to have sunk out of sight at the present time, and which tradition says was to 
the east of Tahiti. — Editob. 



Digitized by 



Google 



TraMtiMS of the Paumatu klan^ 33 

following up the pleasant scent of a '^Atiroro," by which I am 
attracted.) By this figure he meant to typify Huarei. 

** Where is this huraro ? *' inquired Moeava. 

*'Itis at Te Pukamaruia," answered Patira carelessly, without 
even suspecting that by all these words he was rousing the feelings 
of his iuterlocutor. 

Thereupon Moeava retorted drily, ''This huraro tiMragaptmhoiorx^H 
to me. It is already the property of Moeava." 

On one of his previous voyages he had been betrothed to Huarei. 
Patira, on his part, had heard that Huarei was a fair and peerless 
maiden. Her fame had spread even to Marama. But he wished to 
satisfy himself by personal experience, and for that reason was going 
to pay her a visit. 

With one stride Patira was alongside the '' Murihenua," almost 
touching her. Thereupon Moeava, already greatly excited, cried out 
as loudly as he could : '' Off . . . Keep off, or else you will have to 
do with the point of my Puanea ! " Puanea was the name of his 
famous spear. ^ At this threat Patira drew off. But in spite of this 
justified warning of Moeava, Patira none the less continued on his 
way to Napuka. On his arrival at this island he managed to obtain 
an interview with Huarei. She was in very deed a fair and beauteous 
maiden. If ever there was a Polynesian beauty, she was one. 
Assuredly she had not her equal in all the surroimding isles. Her 
beauteous eyes were dear as crystal, her figure was tall and slender. 
Patira was completely charmed, and so, without more ado, he came 
straight to the point. At the end of his interview with her, in order 
to give her a convincing and tangible proof of his love, he stroked 
her cheek. Then addressing her he said, '' Eemain here at home in 
Te Pukamaruia ; I am going away to look after my possessions, and 
in good time I shall return to Te Pukamaruia and marry you." So 
saying be departed homewards, utterly scorning the words which 
Moeava had spoken to him off Kaukura : '' E huraro tuiragapua Huarei 
na Moeava J*^ 

From this time on Huarei was the cause of their mutual bitterness, 
and a little later the cause of the tragic meeting at Makemo. Huarei 
had been betrothed to Moeava from her earliest childhood, and the 
bethrothal ceremony had taken place on Moeava's first voyage to 
Te Pukamaruia. If, therefore, she did not belong exclusively to 
Moeava, at least he had rights with respect to her, which Patira had 
not. Now, the latter, as we have seen, utterly ignored these. 

The ''Marihenua" continued, as usual, to furrow the seas of our 
archipelago, going to the right and to the left, venturing into such 
powerful currents as the Te Koihagaiti, Te Koihaganui, Te Mara- 
tuhakonokono, Te Marapoto, Te Mararoa, etc. Thus he reached 
almost the middle of the famous cuiTcnt Te Moemoe, situated in the 



Digitized by 



Google 



34 



JOURNAL OF THE POLYNESIAM SOCIETY. 



nanow channel (Je gttre\ which separatee Napuka from Te Poto. 
While in this situation Moeava suddenly heard the plaintive cry of a 
bird. He immediately cried out, ** vai ra teie i papaki i U tara o 
Murihentta f (What is it that is thus splashing against the side of the 
" Murihenua " ?) 

It was a small Eupe,^ a kind of pigeon, which was making itself 
heard near the ** Murihenua." As it flew away it repeated over and 
over again, " van teie o Rupe i fane,^^ (It is I, the Rupe, who am 
thus flying about). " I am the Rupe who bathes in the waters of .Te 
Fanomaruia, on my island of Te Pukamaruia" Moeava on looking 
up saw that it was indeed a Rupe. Its cry revealed the beautiful 
voice of a maiden. He was perhaps mistaken about it. Moeava 
addressing it said, ^^E Rupee! e huraro tuiraga pua hoe na Moeava,^* 
The gentle pigeon uttered its plaintive cry once or twice, then swiftly 
fleeing in a trice reached Te Pukamaruia, which was visible on the 
horizon. Upon this significent omen Moeava said to himself, ** I 
shall go and see my little flower that is unfolding herself at Te 
Pukamaruia." Just at that moment tlie breeze sprang up, the sail& 
of the ** Murihenua" filled, and, in a short time, Moeava had the 
happiness of lauding on the island he desired so much. Before 
landing and going to meet his beloved betrothed, Moeava composed 
the following pehe (or song) for the occasion : — 

Ko vau ra ki to moe It is I who am in this place. 

E aha e Rupe ! i fano mai Why then hast thou, Rupe, had to fly 

Teie te Moemoe rihi e ? thitherward over this terrible current of 

Moemoe P 

Ko vau ra, etc. 



Natira (? KatiraJ te moe ! 

E aha ko vau ra ki Te Pukamaruia 

Kahopu, kapahakeo 
E aha ko vau ra ki te vahine 

nui tapairu, Huarei 

te rihi ro mai i. 



Why am I at Te Pukamaruia ? *Tis. 
love ; 'tis the meeting. Why am I here,, 
if 'tis not to see this charming Huarei,. 
who this day is mine for good and all ? 



E aha koti hoki manu iti 
te ipo tagi mai 
teie te kuriri tagi mai ra 
Ko vau ra ki Matiti maru, e ! 
Ko vau ra ki Matiti maru, e ! 
Kahopu, kapahake. 



What dear birdling, 

Means thy plaintive song? 

This is the kuriri which makes its voice 

heard. 
'Tis I, the man of Matiti maru, 
Objects of my love and my desire. 



E aha ko vau ra ki te toa nui, 

Moeava, te rihi ro atu u ; 

E aha koti hoki mapu iti to ipo tag^ 

mai 
Teie te koriri tagi mai o ran e ei ai i i e. 



*Ti8 I Moeava, the great warrior. 
Who have just given up myself wholly ► 
What is there then, dear little friend, 
To make thy calls understood ? 
'Tis the kuriri that raises its voice. 



Digitized by 



Google 



Traditions of tho Paumotu Islands, 35 

Such was the pehe sung by Moeava before going on to embrace 
his beloved Huarei, who dwelt hard by the marae of Eagihoa. With 
her consent previously gained Moeava married her. The bride was a 
direct descendant of Maruia, the first qneen of Napuka. Consequently 
she was herself, it is said, queen of Te Pukamaruia, Te Poto-nui and 
Mahina-te-Tahora. Soon she gave birth to a beautiful little boy, 
whom his father called Kehauri. While giving him this name 
Moeava addressed to him the following words : '* O tei toe icHu net na 
oe ia e amu, e ia pau roa,^^ "All that I shall have left you will devour, 
you will finish." 

Kehauri was endowed with a fresh, ruddy complexion, which 
earned for him the nickname of *'0-ura-noa" (the ever burning). 
From his birth up to early manhood he scarcely ever left his father's 
side. As soon as his son had fully attained his manhood, Moeava 
armed him in knightly fashion with a strong and beautiful spear 
named ** Pakekerua-ki-te-rangi." It was this famous spear which 
dealt the death-blow to the numerous host of Muta, and to the others 
who accompanied him. Thus did it avenge the death of Tagihia and 
his brothers. 

Now, Patira at the time previously arranged, put his plan into 
execution, and came to seek Huarei at Te Pukamaruia. Several 
years after their marriage Moeava had taken his wife back to this 
island, and had left her there with her relatives during one of his 
voyages round the group. Patira therefore found Huarei all alone. 
In vain did she remonstrate and use every means to dissuade him 
from his intention to carry her o£E. Amongst other words she said : 
''Is it possible that you do not know the toa (valiant and powerful) 
Moeava ? " Patira pretended not to know him. **No," said he, ** I 
do not know him." "What!** said Huarei greatly puzzled, "You 
do not know Moeava, the hero, the champion of Vahitu,* of the tribe 
of the Goio-Tuarehu, who hovers with swift flight in the clouds of the 
North? '* Patira persisted all the more, constantly repeating, ** No, 
I do not know him, but what I do know is that I myself am champion 
in Marama, in the tribe of Tokorega." So saying he seized Huarei 
in his arms, by force, and carried her across the seas. 

At that time Moeava was also on the high seas, and by a happy 
chance one day saw Patira crossing at a single stride the channel 
separating Napuka from Katiu, and holding in his grip the 
unfortunate and beloved Huarei more dead than alive. Immediately 
the blood rushed to his head, and he was seized by a frenzy of rage. 
To glut his vengeance Moeava then and there challenged Patira to 
single combat. ** JS ho atu ta taua taniaki i Makemo i te tahua e rca,^^ 
(Come, let lis settle our quaiTel at Makemo-roa-hua.) The wrath of 
Moeava knew no bounds, while Patira was quite willing to accept the 



Digitized by 



Google 



36 JOURNAL OF THE POLYMESIAM SOCfETY. 

challenge. It was therefore arranged that the encounter should take 
place on a fixed date in Makemo, at a place named Te Pohue.^ As a 
matter of fact, there was in this spot an arena eminently suitable for 
jousts and challenges of this kind. The two antagonists were almost 
equal in strength and skill, while thej both enjoyed great renown in 
their respective islands. At the time appointed, true to their engage- 
ment, they an'ived at the place of combat. In spite of Patira's rapid 
means of travel, his antagonist, however, reached the island several 
days before him. As Moeava was sailing along keeping a sharp 
lookout on the channel between Katiu and Makemo, he suddenly saw 
Patira plant one foot on Te Pana. The former was just then in the 
place called Kaumati, while the other foot of the latter was still in the 
sea of Gatika. On seeing his foe Moeava immediately seized his 
magic belt, ** Manava-apoapo," and put it on. Without delay he tore 
from the ground a kind of liana, correctly known by the name of 
pokue, plaited it into a solid cord and made a sling of it. Into this 
sling he placed a hard, smooth stone called, ** Amiomio-i-te-ragi," ^ 
brought in olden times from Tahiti, according to tradition, by 
Hono-ura, who left it at Te Poto on the death of Toarere, a 
Marquesan toa, whom he had conquered. 

Moeava, sling ready in hand, stood erect on the islet of Raumati^ 
scanning/v^ith well-skilled eye the horizon from North to South to 
detect the arrival of his foe Patira. Suddenly he saw a foot planted 
on Ororia. Then Moeava, in order not to neglect his good custom 
taised his war-song. 

It is a prayer to the god Tu. 

Ka hohora i tai eid te heiva O Tu, oome to the combat on the lagoon's- 

E Tu e, e rorei o, marge. 'Tis oalm, oh Tu, dead calm. 

E Tu e, rorogoi ai te matai Oh Tu, brave Tu, come and be present 

Ka hohora i tai eki te heiva at the combat on the lagoon's marge. 

E Tu e, e rorei o, 'Tis calm, oh Tu, dead calm. 

E Tu e, rorogoi ai te matai. 
rau e — ^i — ai — i — ^i — e. 

Ka hohora i tai eki te heiva Come to the combat, Oh Tu, on th& 

E Tu e, rorei o, lagoon's marge. 'Tis calm, oh Tu». 

E Tu e, e rorogoi ai te matai dead calm. Come protect and save fronk 

kia hume. kia hume Moeava- death Moeava. Oh Tu, calm reigns, 
Tukirima noho i ta ora e e ; d^d calm. Come, call on my ancestor 
E Tu e ; e roroi o ; Tapakia. 'Tis calm, oh Tu, dead calm. 

E Tu e, rorogoi ai te matai kia 
heke mai ru ; kia heke mai 
ruga e tupuna ra ko 

Tapakia ; E Tu e, e rorei o 

E Tu e rorogoi ai te matai rau 
e— 1 — ai — i — i— e. 



Digitized by 



Google 



Traditions of iho Paiunoiu h/aMf$. 37 

No sooner had Moeava finished his pehe than he saw Patira 
bringing up his other leg. '^ Now," said he to himself, *^ is the time."^ 
He adjusted the stone in the sling, swung it backwards and forwards 
with ever increasing rapidity, and finally whirled it round his head 
as vigorously as possible. After manoeuvring' thus for several 
minutes, he discharged the stone witli remarkable accuracy. 

Patira was struck fair in the forehead and killed on the spot. The 
stone after having struck Patira rebounded to Retiega, on the edge of 
the lagoon. The result was therefore fatal to Patira, and Moeava as 
conqueror was covered with glory fi*om this fight, which has remained 
famous in the annals of the archipelago. A huge crowd of 
Paumotuans had gatliered to be present at the combat, being anxious 
to learn the issue of a contest in whicli feeling ran so high. By a 
skilful and well calculated turn of his arm Moeava had overthrown 
witli tlie first blow his gigantic and' ten-ible advei-sary, who fell face 
downwards, stretched out at full length. While his head extended 
beyond the outer reef, to the north of the island in the direction of 
Taonga (Tautau), his feet remained bathed in the calm waters of the 
lagoon. The stone which dealt the fatal blow to Patira is still to be 
seen to-day at about a metre's depth, covered with a young growth of 
coral, in the clear water of the lagoon off Makemo. 

At the fall of Patira there went up from the excited crowd of 
onlookers a mighty shout of joy, accompanied by endless cries of 
*'Huro" and frantic applause. Moeava immediately rushed to his 
victim, and, spear in hand, ran along the full length of his body from 
foot to head, as though he were crossing a bridge, and With his 
miglity Puauea he cut off his head. With one hand he seized the head 
of Patira by the hair, and with the other he released his beloved 
Huarei fi-om the gi'ip of his vanquished foe. Both these he carried 
to the interior of the island, where he at once prepared the oven for 
his enemy. Patira's head he gave to Kehauri. Thus having satisfied 
his vengeance, he returned to Takapua in a happy frame of mind^ 
taking with him his wife and son. He had never yet been parted from 
the latter. 

It is clear, and there is not a shadow of doubt about the matter^ 
that Huarei was the true and only cause of the strange combat 
between Moeava and Patira, the combat in which the latter was slain. 

On the other hand the death of Patira was soon avenged by his 
fellow-countrymen, the numerous tribes from Marama, and especially 
that of Muta. They made an inroad upon Takaroa, and, not finding 
Moeava there, avenged themselves on his nephews, who had become 
his adopted sons. 

Moeava, in turn, was not slow to exact vengeance for the deaths of 
Tagihia, Parepare and Rogotama, by captunng at Punaruku 
(Makemo) the host of Muta. It was certainly not jealousy on the part 



Digitized by 



Google 



38 JOURNAL OF THE POLYMESiAN SOCIETY. 

of Kehauri about the turtle's head, that caused the tragic death of 
Tagihia-ariki, and of his two brothers, as some liave thought. As a 
matter of fact, Kahauri had uo intercourse, still less any consanguinity, 
with these tribes from Marama. On the side of his father Moeava he 
was from Vahitu, of the tribe of Goio-Tuarehu, who originally came 
from Takapua and Haoroagai, aud on his mother's side he belonged 
to the tribe of Quti-Hopu (Napuka). 

Later on, after the conversion of the natives to Christianity, and 
after they had gained u knowledge of the Bible, they did not fail to 
compare the strange combat between Moeava and Patira with that 
between David and Goliath. There is indeed a striking resemblance 
between the two. 

NOTES. 

1 . The following is a brief description of this canoe : — 

The outrigger was called Oheohe 

The bow '* kiato " was called Hotutaihenuku 

The stern ** kiato " was called Paratito 

The steering paddle was called Taripo 

The mast was called Tiriatofa 

The sail was called Kukuti ki te ragi . 

The forward thwart was called Kifakatakuariki 

The aft thwart was called Tearokaharia 

The end of the mast was called Kifaretataha 

The extreme end of the sail was called Kitaiomere 

The ** tatakoto" (boom of sail) was called Tiriakoukou 

The extreme end of the ** tatakoto " was called Fani 

The forward cabin was called Te piha tuatiaki 

2. The spear in Polynesian lands is the emblem of worth and 
courage. The spear of Parepare was called Te Arovaru, and that of 
Bogotama, Terefa. 

3. I am informed that this bird was at one time found at Te Poto. 
To-day it is found nowhere except at Makatea ; elsewhere it is 
unknown in the Tuamotu Group. 

4. Vahitu, of which Tufaruia was king, is the ancient name of 
the second group lying to the West of the Tuamotu. It comprised 
the following nine islands : Aho, Manila, Takai'oa, Takapoto, Tikei, 
Taiaro, Aratika, Kauehu, Earaka. 

5. Tahua was the name of this region of the island, and its king 
was Manumea. The marae was called Te Utuga, the lagoon in this 
place Waihunu, and the tahora, 6 Taganiui ia. 

6. According to Piritua, the present chief of Makemo, the name of 
this stone was *' Pohatu taka i marama,^* 

7. Tu was one of the gods of Moeava. But his chief divinity was 
the god Tagigorigo. For Tu see Mangarevan -French Dictionary, p. 
Ill, under Tu, and also this " Journal," Vol. XXVIL, p. 124. 



Digitized by 



Google 



THE SAN CRISTOVAL HEO. 



By the Rev. C. E. Fox of San Cbistoval. 



THOUGH the islai^ of San CristoveJ is not large — about eighty- 
miles long and twenty miles broad — and has now only a 
population of about 8000, a great many methods of disposal of the 
dead are found on it. More than twenty- different metlK)dsajpe known 
to me, but some of tliese differ little from one another. Burial on the 
A^ is only one of these. 

The following sketch of the heo at Ubuna is by a native, and I 
have not seen this particidai* one. 




>v/ 

(a) Herat an open space ; usually the namQ for the village burial plaoe^ 

also for the open square of the village. 

(b) Heo, of earth and stones. 

(e) Hau 8WU, a stone receptacle for the bones of the dead, made of five 
large stones, forming the four sides and top. 

The heo (of which there are many score) are not usually large, the 
largest being perliaps twenty feet high and thirty feet square. Two 
large ones were lately levelled at Wango, and the bones, of which 
there were a gi-eat number, removed. All these bones showed burial 
in the horizontal extended position. 

The heo sometimes belonged to particular families of chiefs, and so 
far as I know at present only people of the chief's clan were buried 
on them. A hollow was made on the top of the heo, and in this hollow 
0, platform, on .which tlie dead man was laid, l>ound up in pandanus 
mats. The corpse was then taken at frequent intervals to running 
water ; it was carried pick-a-back by a man called "the keeper of 



Digitized by 



Google 



40 JOURNAL OF THE POLYHESIAK SOCIETY, 

the dead " ; when lie reached the stream he crouched down, with the 
dead niau still on his shoulder, and water was poured over the corpse 
from the stream, after which it was carried hack to the heo. ** Tlie 
keeper of the dead *' had a very unpleasant time and usually could 
not eat for some days, and did not speak until then. His head was 
protected hy a native hag. In some parts, however, water was 
brought to the h^ and there poured over the corpse, tiU the flesh came 
away, when the bones were buried on the heo, or put in the hau sum 
(all the heo did not possess a hau suru). Over the platform on .the heo 
a roof of sago palm thatch was made. 

The heo are found all over the western jyart of the island from 
Wan go on the north coast round to Makira Harbour on the south 
eoast. They are usually of earth, flanked with stones, and the ones I 
have seen are more or less rounded, with flat tops, oblong in shape, 
but hardly suggesting a pyramid, as the sketch does. 

It seems to me that the maeitawa of Haununu on the south coast 
(to the east of the district where the heo are found) throw light on the 
heo. The masttawa (a word that in neighbouring parts means iKMtt 
harbour^ are' the burial places of that part — circular cleared spaces. 
Common people are buried round the circumference, facing outwards, 
lying horizontally. Chiefs are buried in the centre of the maeiiaway 
usually at least facing east (a very unusual position in other parts of 
the island). A hollow is made about thirty or forty feet long and 
twenty feet broad and in this a hous^ is built, rather different, how- 
ever, from an ordinary native house, as two poies are fixed crosswise 
at each end x — ^x, a ridge pole placed on top, and sloping sides and 
ends of sago palm added, like a sago palm tent. An opening is left, 
flo that the jawbone of the chief may be taken when the body on the 
platform, within the house, decays. The whole is then covered over 
with earth, and large stones placed along the sides. After a time the 
whole falls in making a broad, low mound, only a few inches high, 
flanked by stones, and such are the only masitaiva 1 have yet seen, in 
each of which only one cliief was buried ; but if, as I am told is the 
•case, a number of chiefs are buried in succession in tlie same maeitawa^ 
one would expect a mound similar to the western ^ to be the result 

The wumiawa themselves seem to be modifications of the burial 
^daoes inland, which are circular cleared spaces, of about the same 
aijEe ; but in these a sacred tree is found in the oentre instead of the 
chief's mound, and the dead are biuried fairly deep, in a sitting 
position, securely tied, in concentric circles facing tlie sacred tree, 
which is vailed '' the viUage of the dead." 

The heo of the western district differ from the mmniatca mounds, 
in that they are always at the west end of the htra^ which is not 
circular like the moiiUuca^ but oblong. 



Digitized by 



Google 



The San Cnsio¥al Heo. 41 

The chief other methods of disposal of the dead, on San Cristoval, 
are: (1) burial in the sea in either upright or sitting position, (2) 
cremation, (3) laying out in extended position in large bowls or on 
platforms till the flesh decays, (4) embalming in canoes called ''the 
canoe of the sky," (5) laying out on crags and pinnacles of rocks 
along the coast, in extended position, (6) partial burial up to waist in 
upright position. But there are many modifications of these different 
methods. 

The note on the heo and masitawa must be considered as preliminary 
only, as I have not yet sufficiently examined them. 



Digitized by 



Google 



TRADITIONS AND LEGENDS. 



COLLECTED FROM THE NATIVES OF MURIHIKU. 

(Southland, New Zeaijind.) 



By H. Beattie. 



PART IX. 
Contintied frmn Volume XXVII., page 137. 

In gathering the material for these articles the Collector has secured 
a large number of notes which, while meagre in themselves, are yet 
worthy of record both because of their intrinsic value and because 
they may put historians on the track of further information. 
These notes are herewith presented. 

FIRST FIRES. 

In an old exercise book (of which the first page had most 
unfortunately been torn out and lost) on page 2, there were the 
following two notes : — **iVa Pou t whakaha te ahi hi Tawhiti-nui, ka 
PoU'tama-nui te kauati ; " and " Na Tnra i whaka {? whakakd) te aht 
ki Tawhiti'mii-a-Rua \ ko Mata'aitu te kauati \ ko Tauira-o-hua te 
tamaiti a Twa^ 

These notes signify that Pou was the man who first lit a fire in 
the land called Tawhiti-nui, and that the name of his firestick was- 
Poutamanui ; and that Tura was the first man who lit a fire in the 
land called Tawhiti-nui-a-Rua, his firestick being named Mata-aitu,. 
and his son was Tauira-o-Hua. 

Mr. S. Percy Smith supplies the following : — "Pou is supposed to- 
have been carried away to Tahiti by a big tantwha, and a bird brought 
him back. The Tawhiti-nui mentioned is probably the island of 
Borneo while Tawhiti-nui-a-Eua is Tahiti Island. Rua after whom 
it is named lived there — he is known traditionally to the present 
Tahitians." 

The collector asked Tare te Maiharoa regarding the islands known 
to the Maoris and who lit the first fires on them, but he replied he 
had never memorised the list. The first page of the manuscript book, 
which had been lost, had qontained a list of the islands and the first 



Digitized by 



Google 



Traditions and Legends. 43 

men to build hearths on them. Tliei-e was not room to write all on 
the one page so two entries had been placed at the top of the second 
page. This page has been preserved ; the other went astray many 
years ago and no one can now recite the list. This is the tragedy of 
this information not having been collected thirty or forty years ago. 
(My informant added he did not know the name of Pou's canoe, but 
Tura went from one land to the other on a rainbow, whose particular 
name he forgot.*) 

AN ANCIENT KARAKIA 

The following is a very ancient karakia, but the collector has no 
particulars of its origin or use : — 
Mo Eaki 
KiaTuhaha 
Kia Tumoremore 
Kia Tu Tahaka 
Kia Tu Koauanake 
Ko Te Kapua 
Eo Te Mihaia 

CREATION OF MAN. 

One or two of the Waitaha genealogies, copied by the collector, 
were prefaced by the following remarks — " A genealogical account 
of the descendants of Tane who created man from the earth. Tiki 
was the first, and the second created by the same process was lo, a 
female, whom Tane gave as a wife to Tiki, and tliey * poured out ' 
mankind and filled the earth." 

One of my informants said : — " Tane, the god, at tlie beginning 
of the world, married Hine-ti-tama and had two sons named Tahu- 
kumea and Tahu-whakairo. When these two died Tane went to 
another land and there he created mankind, Tiki being the first man, 
and lo the first woman. Tane altogether had seven wives of whom 
six bore issue. From the third wife, named Hotupapa, there is a 
very long whakapapa which comes down to the Eati-Mamoe tribe, 
but I cannot repeat it. From the fifth wife of Raki (? Tane) the 
descent comes down to Matiti and the Waitaha tribe. The men of 
knowledge among the Maoris forty or fifty years ago could have 
recited all those things but they are now forgotten." 

A CELEBRATED AXE. 

In a notebook written by Wi Pokuku in 1880, occurs the following 
karakia accompanied by a short account concerning tlie axe used by the 

* Tura is a well-known ancestor of the Maoris and flourished in the time of 
Whiro, or twenty-six generations ago. Apparently he lived in either Ra'iatea or 
Porapora Islands of the Society Group.— Editoe. 



Digitized by 



Google 



44 JOURNAL OF THE POLYMESiAM SOCIETY, 

famous Bata to fell the tree for his oanoe. The name of the axe was 
' Aumapu,' and its karakia was : — 

£ atno to toki e aku tanka 

E wa (P waha) ki te tamu o te rakau e, 

E aku tauka ko 'Aumapu ' ma te toki 

E uru taku loki e uru ki tokerau 

E aka taku toki e aka ki te uru o te ra 

E tua taku toki o tua ki te uru o te tonga 

E tere taku toki e tere ki te ua o te raki 

Tere taku toki e, tere taku toki 

Tere taku toki e, tere taku toki 
Uruuru atu taku toki ki te tua o te ra 

Tere taku toki e, tere taku toki 
, Tere taku toki e, tere taku toki. 

Koia net te karakia o taua toki, o ^ Aumapu,^ kia mohio ka takata ko 
te toki nana a * Takitimu ' i tarai ko * Aumapu,^ i a Rata tena toki e 
takoto ana, na Kahue i tuku mai ki a Rata. Ko Rata ki te tapahii taua 
rakau nei, te ra kei te whawhai a Ruru, mahara ki to te moana takata, 
Ka mate a Ruru ka lohakarauoratia e Rata a Ruru, no muri mai ka 
haere a Rata ki te kimi rakau hei ivaka mo ona kai kaki i te mate o tona 
hakoro, o Wahieroa, 

Notes. — Rata according to the tables in ** Memoirs of the Poly- 
nesian Society," Volume lY., lived 39 generations ago. (See page 
234 — the table facing page 120 is numbered eight too many right 
through, and makes Tamatea 30 generations ago instead of 22). He 
(Rata) was the contemporary of Kupe and Ngahue. When 
Rata wished to fell a tree to make a canoe to go in search of the 
people of Matuku who had killed his father, Wahieroa, he went to 
Ngahue (or Kahue) and asked for an axe. Ngahue broke a slab of 
stone and made three axes — one for himself named * Kapakitua,' one 
for Kupe named *'Tauira-a-pa,' and one for Rata called *Te-papa- 
ariari.' Rata sharpened his axehead, put a handle on it, and named 
it * Aumapu.' (White's " Ancient History of the Maori," Vol. t, p. 
74, gives this name as * Mapu-naiere.') With this axe Rata went into 
the forest to cut a tree — one of the most frequently told incidents in 
Maori liistory. 

RATA AND RURU. 

In all the many vei*sions of the story of Rata the collector has 
never noticed the name of Ruru, except in the fragmentary account 
given by Wi Pokuku just recorded. 

Asking after Ruru, the collector was told that Ruru was the king 
of the small bush birds, even as Totara is king of the forest trees. In 
the days of Rata the bush birds fought the sea birds, and were defeated 
and fled into the bush. Rata saved the life of their king Ruru, and 



Digitized by 



Google 



Traditions and Legends. 4$ 

in return for this kindness the latter helped Rata to get a canoe. 
Twice the tree that Rata liad cut during the day was erected during 
the night by the small birds and the leaves of the forest. He watched 
the third night and saw this done. Rum appeared and told him the 
reason the tree he had cut had been re-erected was because he had said 
no karakia prior to his work. Ruru taught him the karukia to say and 
told him to go hpme. In the morning he returned to find the bush 
birds, under Ruru, had hollowed out a fine canoe for him. He named 
tlie canoe ^Tarai-po' because it hud been chipped out in one night. 
This caiioe was afterwards re-named *• Taki-timu ' (in memoiy of the 
stump of the tree from which it was cut), and voyaged about a good 
deal. With the aid of Te-Tiui-o-te-para-rakau people, Rata 
accomplished his revenge on the Matuku people for the murder of his 
fatlier. 

(Note. — The * Taki-timu' canoe mentioned is evidently not the 
famous one of the same name which came to New Zealand. The name 
of the canoe of Ratp. was also given to the- collector as*Niwai*u.' 
Another old man said, "Tinitini-o-te-pararakau are now the evil 
spirits of the trees, and if you do not karakia to them will hinder your 
tree-felling. They sometimes appeared as birds, and in this foim they 
helped Rata.") 

[The story of Bata and the Ruru is well-known to the Rarotongans, and will 
appear in great detail when the series of Rarotongan traditions are printed in. 
our pages. The scene of the story is laid in the Samoa Islands. A brief 
aocouut of Ratals adyeu tares appeared in '* Journal Polynesian Soeietyt'^ 
Vol. XXI., p. 61.— Editob.] 

A CATALOGUE OF FIGHTS. 

The collector came across two lists of fights compiled from 
Kai-tahu and Kati-mamoe sources, and herewith presents the names 
with such brief details as he could gather : — 

I. Te Kiore-mauhope, fought in North Island. No details 
available. 

a. Rauwhata (Nortti Island). Here Tu-kake-mauka and Te 
Whatu-kaipapai were killed. They died without issue. They were 
tlie sons of Tn-maikuku, by his wife Irakukuini, and were grandsons of 
Hikaororoa and Urupa. 

3. Whata-roa at Tnraka (Poverty Bay). Rakatoatoa and 
Manumai the only chiefs of the defeated side who got away, escaped 
in the smoke. This fight is also known sometimes as Kai-whakaware» 

4. Te Kai-whakatari, at Poverty Bay, is said to have been 
fought over a dog. Kurawhaia an ancestor of Tu-te-kawa was killed 
there. 

5. Te Kakihaua fought in Noi-th Island. No details. 

6. Hikaororoa fought in North Island. No details. 



Digitized by 



Google 



46 JOURNAL OF THE POLYMESIAM SOCIETY. 

7. Marukere fought in North Island. No details. 

8. Te Piki-tU-roa. Hei© Marukoi-e was to be cooked, but his 
youngedt brother Hoko-pae-kawa put on Mamkore's piki (pluuie), and 
took his place for the oven. The plume stuck out of the oven (a bad 
omen) and the body did not cook properly so was cast away. Hencd 
the name Piki-tu-roa (long-standing plume). 

9. Te Kaihuka (North Island). Pi to and Baukawa were killed 
here. 

10. Hukete. Here Huirapa and Te Maraeroa (brothers) were 
killed. They were tapapatajca (lying down) when killed. 

11. Tapapanui (also called Tapapa-ruahine). The Maori text 
is translated by Mr. S. Percy Smith as : " The deaths of Huirapa and 
Te Maraeroa were avenged by their younger brother Tahnmata, and 
hence came the tribe Te-aitaka-o-Riti to this island in order that 
they might incorporate with Te-aitaka-o-Tapuiti and Waitaha tribes 
to save themselves (from extermination?), but Waitaha would not 
consent ; then turned on them and avenged (the deaths) in this island." 

13. Waipapa. Here Kai-Tahu beat Kati-Mamoe. It is in 
Marlborough. The Kati-Mamoe were caught unawares cariying 
loads of fern -root on their backs and were vanquished. 

13. Te-ika-a-Whaturoa. " Up Kaikoura way." No details. 

14. Huriwai. Here Tuhuku the Kati-Kuri chief was killed by 
the Kati-Mamoe who won — it was a canoe fight. 

15. Pauariki (also Pouariki). The Kai-Tahu chief Tahu-tutua 
was killed here. 

16. Puhirau. Bakaimomona, a Kati-Mamoe chief and father of 
Tukiauau, was sitting with his back to the sun, sunning himself 
outside the Opokihi pa, when Manawa, a Kai-Tahu chief, from an 
adjacent cliff threw a spear, which killed him. 

17. Opokihi. Here Bakai-tauheke killed Popoia and Tarere, 
while Maru killed Te Puehu and Te Aweawe. Bakai-tauheke was a 
man of gigantic strength and killed his purua (pair) with one blow of 
liis ^ann (a black stone weapon held like a mere). The weapon not 
only killed the two victims but stuck in the karaka tree behind where 
it is said its embedded head can still be seen. The Opokihi pa 
contained Kati-Kina, Bakitane and Kati-Mamoe people and fell befoi-e 
the invaders. 

18. Peketa. Kai-Tahu beat Kati-Mamoe. 

19. Te Pari-whakatau. Manawa was killed by Tukiauau, and 
honoi's were even. The account briefly says " te otika te riri " (the 
end of the war). It is in the Kaikoura district. 

ao. Teihoka fought near Colac Bay. Kati-Mamoe administei-ed 
a decisive drubbing to Kai-Tahu and saved their tiibal identity. 

31. Taupirl was a pa on the northern side of the Hokanui Hills, 



Digitized by 



Google 



Traditions and Legsndi. 47 

and was evacuated by the Kati-Mamoe before a Kai-Tabu taua. Some 
people were killed here at a later date. 

33. Tarahau-kapiti (also^known as Wai-taramea) was fought 
on the Five Rivera Plain. One old man said : " Elaweriri grabbed an 
old Kati-Mamoe chief Te Kairere, and also the latter's nephew 
Tu-te-makohu. He killed the uncle, but the nephew killed him. The 
heart of Te Kairere was roasted and eaten." Anotiier said that 
though Te Kairere was killed he was doubtful if Kaweriri personally 
killed him, and although some of the hearts of the slain were eaten, he 
never heard that Te Kairere's was. The Kai-Taliu won the fight, but 
disheartened by the death of so many of their band (including Kawerin 
their leader) they returned to Oantefbnry, and Tu-te-makohu dwelt 
peacefully at Otaupivi until his death, 

33. Otauaki was fought in Westland. There were a few words 
in Maori against this name, and Mr. 8. Percy Smith wi'ites : — " This 
note is interesting as it gives tlie name of one whom we may call * the 
warden of the marches.' The translation is : Tauaki gave birth to 
Hauaki who was the great chief of Patea. He was the 'shield^ 
against Poutiui (i.e., the West Coast Maoiis)." A further note, 
partially undecipherable, says that * Te-aitaka-a-Riti ' and * Te- 
aitaka-a-Tnpu-iti ' had something to do with the fight. The collector 
Was told that the people known as Patea were a combination of Rapu- 
wai and Kati-Mamoe. They were attacked by Kati-Mamoe from 
another part, led by Takai-waho, and were defeated, Hauaki being 
killed. 

34. Hunoa. This was in the time of the Kaihuaka troubles on 
Banks Peninsula, but the collector has no details. 

35. Kai-huaka (" eat relatives ") is the name of a feud which 
occurred between sections of the Kai-Tahu tribe shortly before the 
White people came to Canterbury. The collector was told the name 
originated through the starving people in a pa eating each other's 
children, and again he was informed that in the fighting a number 
who were killed had their heads cut off, and the bodies were eaten by 
relatives unawares. 

36. Kai-whareatua is said to be the name of a southern foray 
against the North Islanders in Te Rauparaha's time. 

37. Taua-iti is the name of a raid made by the Kai-Tahu and 
Kati-Mamoe against the Kati-Toa in Marlborough. 

28. Tau-whare-kura was a fight between Manuhune and 
Takapo in the lake district at a hill called Mt. Grey by the Europeans. 
Kati-Kuri built a fortification there, but the collector has no further 
details. 

39. Oteihokawas a fairly recent fight in Canterbury. A section 
of a tribe known as Ghapuku was chased by a composite brigade of 



Digitized by 



Google 



48 JOURNAL OF THE POLYMESIAHI SOCIETY, 

Waitalia, Kati-Mainoe and Kai-Taliii descent to Te-Umu-kaha, and 
satisfaction obtained at the Harakeke-tau-^toru pa there. 

30. Taiarl. Here Tu-takabi-kura was killed by Tu*te-inakobu. 

31. Ohinekete on the Otago Peninsula, was where the Kai-Tahu 
chief Tane*wliaka-toro-tika was killed by Te Maui, a Kati-Mamoe 
chief. 

3a. Ohou was a fight at the lake of that name. The notes briefly 
«ay, *'Te Kainiutu was killed by Te Bakitauhopu," but the collector 
was told : — ** Pohowera, a Kati-Mamoe chief, was killed by two 
Kai-Tahu chiefs Te Kaiuiutu and Tawhiriruru. Pohowera's sou Te 
Hahitauhope (my infoimant was positive this name finished with an e 
not u) killed these two chiefs at Ohou in revenge. Then another taua 
went up there and Te Bakitauhope was killed by Kaunia." 

33* Katiki. ^' Te Matauira a famous Kai-Tahu chief was killed 
at Katiki (now known as Kartigi). He was told to say karakia before 
entering the pa. The war was caused by Taoka killing someone up 
North. Te Matauira was killed as follows : A man was i-unning round 
a whore, and another man threw a spear over the house, and it landed 
on Te Matauira. Other killings followed this." 

Notes. — This list of fights known in Southern history is not in 
chronological order by any means, and is very incomplete as the 
collector could name off-hand half-a-dozen fights that are not mentioned, 
and that were more important than many' on the list. It is valuable, 
however, as it records several names that are new to the collector, and 
about which little appears to be known. 

The collector has further notes to the effect that Hapai-ki-waho 
was killed at Hakaia by Tu-manihi and Tu-te-kaehe (?), and that his 
•death was avenged by Hikatutae and Te Ariki (?), and that further 
fighting took place at Arowhenna where Taka-ahi and Hapi were 
killed, but his notes require corroboration and extention. A further 
note states that the natives at Te Muka built a fortification at Wai-a- 
te-ruati when Te Bauparaha took Kaiapohia, but the collector has no 
further details of it. This was the last of the memoranda jotted down 
in the Maori notebooks and copied by the collector, but he has details 
of fights not mentioned in the list and not hitherto recorded and these 
will be given later. 

MULTUM IN PABVO. 

The collector has a large selection of miscellaneous jottings, a 
number of which he will give here with the hope that they may put 
those engaged in research work on the track of fresh information on 
flome of these points : — 

Pipiwharauroa. This bird, " the shining cuckoo," is called Te- 
manu-a-Maui in Murihiku, because its song repeats some of the wordfi 



Digitized by 



Google 



Traditions and legenih. 49 

«uiig by Maui when in the fonn of a bird on the handle of his father's 
Jco (spade). 

Te-Tau-a-Tarawhata was a canoe belonging to Waitai when the 
Kati-Kiiri invaded the South Island. Waitai came down to Otago in 
this canoe, which was afterwards dashed to pieces at the foot of the 
lofty cliff now known as Te-tau-a-Tarawhata. Tlie railway line 
north of Port Chalmers passes round this cliff. 

Te-tatau-o-te-whare-o-Maui (the door of the house of Maui) is 
the Munhiku name for the insect known as Daddy-long-legs. The 
name has something to do with the story of Maui, but the collector 
could not ascertain wliat. 

Uruuru-whenua. At the head of the Hakaia River near Mt. 
Somers is a sacred tawai (beech) tree standing by itself. It was used as 
an uruurU'whenuaj and was reverently approached by the people who 
laid offerings of food and other things before it. 

Papapuni. According to tradition Tukete, a chief at Rakiura 
(Stewart Island) was prodigiously fat. It is said he suffered fix>m 
papapuni, a stoppage of the bowels that permitted nothing to pass 
through him. He was killed at the fall of the pa at Putatara. 

Kauheke. Some of the men of old wore the kauheke or chaplets of 
ribbon wood. Chaplets were made of toatoa or celery-top pine. The 
wearing of tliis ornamentation was a tohu rakatira (a sign of 
chieftainship). 

Te kai. In the south food was sometimes eaten only once a day, 
but as a general inile there were two meals a day, the morning one 
being called kai-mo-te-ata, and the afternoon one kat-mo-te-ahi. 

Piopiotahi is the Maori name of Milford Somid. It is said it was 
called after a canoe which came from Hawaiki. Kahotea was the 
captain, and Tangiwai one of the crew, and these two names are now 
applied to different kinds of greenstone. 

Herekopare is an island off Stewart Island. When the Waitaha 
:first visited this locality they landed on tlie island, a chieftainess named 
Mahihi stepping ashore first. Prior to doing so she bound her hair-fillet, 
and the island out of courtesy to her was named "Te-hereka-o-te-kopare- 
o-Maliihi," now abbreviated into Herekopare. It is said there is on the 
island a hill with a mound on it which perpetuates the shape of her 
head and it is called Te Tihi ('' the summit " or top). 

Taniatea was the captain of 'Takitimu' and lived at Tarahaukapiti 
at the foot of Mt. Takitimu for a time. This was a pa-kakart of 
Waitaha. The rua (pit) was there where his people got the karehu 
(soot) for the moko (tattooing). " Te whakatakaka o te karehu o Ta- 
matea" is a proverbial saying. 

Tono te kararere [sic~\ was to send word by a messenger. Sometimes 
when not convenient to send anyone and the wind was in the right 
direction a quantity of dry toe -grass would be scraped and flattened 



Digitized by 



Google 



50 JOURNAL OF THE POLYMESiAIII SOCIETY, 

into a hoop and let go to be carried by the vnnd to the other village* 
Such was the statement made to the collector. 

Te Mu and Te Weka are two old names given to the pakeha. 
Matiaha Tiramorehu mentions them in his Waiata in the " forties." 

Huatapu is well-known in the south. The name of the canoe of 
which he drew the plug was *Tu-te-pewha-raki,' and his club, was 
named * Kahutia-te-raki.' There is an extremely lengthy song about 
this affair* Extra big waves round the Murihiku coasts are. still 
called * Rua^tapu • in memory of him.* 

Kokopu is the New Zealand trout. There is a proverbial sayings 
" jf te Jcakopu te Icai a Mdui,^^ but what originated it the collector 
cannot say. 

Harakeke, the flax, was put to many uses, but probably none more 
novel than signalling. My informant said big wliite mats were made 
of it, and when visitors wei^e leaving a kaika^ or village, these mats 
would be spread oiit on a hill-side facing the next village to let people 
there know of and to expect the travellers. My informant cited the 
case of ,the kaikas at Tuturaii and Hokanui, which had a regular 
system of this method of signalling. - 

Waiariki, or Waitapu, is said to have been the Maoii name for the 
hot springs at Haumer. The Maoris went there for matekohi and 
other troubles, and it is said tliere was a special pool for those under 



Mataehu caused a great flood, said one of my informants, which 
nearly covered the South Island. It was before Rakai-hautu's time,^ 
and it had the effect, of thinning out the Moas, which had been too 
numerous before then. It decimated them almost to the point of 
extinction. 

Te-ahi-a-Ue is a name applied to lignite burning in the ground^ 
Fires of this kind last many years, and have occurred at Te Muka,. 
Waihao Downs, Pomahaka and elsewhere. The first fire of this sort 
was ignited by a man named Ue many centuries ago. 

The Kumara was brought to New Zealand by the canoe 
* Arai-te-uru,' which arrived at Whitiaka-te-ra (east coast of North 
Island), and landed some of its cargo. A storm arose, and it ran to- 
Matakaea (Otago), where it was wrecked and its cargo turned into 
boulders, but the kumaras were not all lost as the Kahui-Roko people 
at Whitiaka-te-ra made good use Of the ones landed there. . 



* Rather m memory of ' Te tai o Ruatapu/ a well-known flood (due probably 
to a hurricane j that took place in Rarotonga not long before the. fleet sailed for 
New Zealand in about .1350, and of which Riiatapu gave warning. — Editob. 



Digitized by 



Google 



Traditions and Legends. 51 

Whetu. Among the southern names for stars are two which the 
collector has not seen recorded by White, Tregear and others, and these 
are the coustillations Kahuiwhetu and Whakarepu-karehu. 

The collector still has an accumulation of material relative to 
southern histoiy and folk-lore which has never been in print, and 
some of this will be given in the next instalment. 

f'To be continued J 



Digitized by 



Google 



THE SCIENCE CONGRESS, CHRISTCHURCH, 
NEW ZEALAND. 

(Communicated.) 



THE Science Congress, which was opened at Christchurch in the 
first week of February, marks a new departure in tlie history of 
science in New Zealand; a departure which it is hoped will liave 
considerable results as far as etiinology is concerned. The congress- 
affords an opportunity which lias never existed before of periodical 
meetings and discussions at intervals of two years. It is hoped that at 
future Congresses anthropology may be constituted into a separate^ 
section; at the Christchurch Congi'ess it was included in Section D,. 
General. Archdeacon Herbert Williams, M.A.^ was elected chairman 
of the section. 

The only papers relating to Polynesian ethnology were those of 
Mr. H. D. Skinner on the material culture of the Morioris, and of 
Archdeacon Williams on the Morion language. 

In the former paper traditions preserved by Alexander Shand were 
quoted as showing that some, at any rate, of the Morion ancestors 
came to the Chathams more than seven centuries ago from 9. land there 
was no difficulty in recognising as New Zealand. As the Chatham^ 
produced no trees large enough for canoes to be made from them, the 
Moriori claim to isolation might safely be allowed. The special 
interest of Moriori material culture lay in the light it threw on the 
history of Maori material culture and art. In the writer's opinion New- 
Zealand, in pre-European times, was divided into two culture-regiona 
whose boundaries coincided in a general way with those of the twa 
islands. Between these two regions was a broad intermediate area 
where the two cultures blended. This division into two areas was^ 
based solely on the evidence of material culture and art, but it was- 
believed that a considerable amount of evidence in support might in 
future be drawn from the study of Maori dialects, and perhaps from 
anthropometric investigations. Moriori material culture might be 
looked on as a. fragment of the Southern Culture of New Zealand. In 
one or two respects — for example, the peculiar wash -through boat— it 
had developed features of the parent culture until, to a superficial view^ 
they appeared entirely new. But even these could be traced to a New- 
Zealand source, while in very many classes of manufactured article it 



Digitized by 



Google 



The Science Congress, Christchunch, M,Z. 53 

was not po9§ible to distinguish the Chatham Island article from that 
made in Otago. This was conclusive evidence that seven centuries 
ago, when the Morioris left New Zealand, the Southern Culture of 
New Zealand had developed most of the features it presented at the 
beginning of the nineteenth century ; in other words, that the 
difference between the culture -regions in New Zealand was ancient. It 
seemed probable that future investigations would indicate a relationship 
between the Northern Culture and tiiat of the Western Pacific^ 
especially of the Solomons and the islands and coasts north-west of 
them. The Southern Culture seemed to find its nearest relationship in 
the material culture of Easter Island. It seemed probable that that 
culture had been obliterated in intermediate islands by later ethnic 
waves and by the influence of intercommunication. 

Archdeacon Williams bnefly sketched Morion traditional history* 
Our knowledge of the language was due almost entirely to the late 
Alexander Shand, who published a series of papers in the " Journal of 
the Polynesian Society," and later compiled a vocabulary which it was- 
hoped might shortly be published. A study of tliese materials revealed 
the fact that the language differed in grammar and vocabulary very 
greatly from the Maori language. Words were altered materially in 
form, and differed in meaning from their Maoii cognates, while about 
ten per cent, were derived from roots no longer preserved in Maori » 
These foreign words exhibited relationships with the languages of the 
Marquesas, Tahiti, Samoa, Hawaii, Tikopia, Easter Island, Mangareva,. 
Tonga, Uvea, Futuna and Earotonga, the likeness being closest in the 
case of the Marquesas, and diminishing in that order. These divergences 
from Maori were th^ more important when it was remembered that 
the Moriori people had been under Maori dominance for over thirty 
years when Mr. Shand began his investigations. As it was the 
language was further removed from Maori than the dialects of 
Barotonga, Tahiti, Uvea, Nine, and could not be designated, as- 
formerly, as a mere sub- dialect of Maori. It could not, however, be 
claimed that the language afforded reliable positive indications of the 
original home of the Moriori, though it was remarkable that his 
investigation and that of Mr. Skinner, conducted independently, by 
different methods, and on wholly different material, should both point 
towards Eastern Polynesia. 



Digitized by 



Google 



NOTES AND QUERIES. 



The Teaching of Ethnology. 

Mr. H. D. Skiuner, B.A., D.C.M., has been appcnnted lecturer in Ethnology 
at the University of Otago, and assistant curator of the University Museum. 
During the session of 1919, au introductory series of lectures will be delivered by 
Dr. W. B. Benham, F.R.S., Dr. Benson, Mr. Skinner, and Dr. Dunlop. In the 
loUowing seasioQ the syllabus for the Diploma in Anthrc^logy will be put into 
q[>eration. The Senate of the University of New Zealand has decided to institute 
a Diploma of Anthropology, and has asked the'Otago committee to report on the 
syllabus. 



PROCEEDINGS. 
POLYNESIAN SOCIETY. 



Thb Council met on the 26th March at the Hempton Rooms, when current 
business was conducted, including correspondence. 

New Members : — 

Mrs. Edith Nairn, Oteka, Havelock North. 
Miss Olive Nairn, Oteka, Havelock North. 
Charles Davis Lightband, New Plymouth. 
Richard Ormsby, P.O. Box 99, Te Kuiti. 
Charies Arthur Budge, Stratford. 

The f ollowTng papers have been received : 

Fakahina Island, Panmotu. By Rev. H. Audran. 
History and Traditions of Rarotonga, part VI. 
The Period of Hiro and Hono'ura. By A. Leverd. 
The Fatherland of the Polynesians. 
Rangi-hna-moa. By G-. G-raham. 
The Science Congress, Christchurch. 

The death of Mr. J. P. Cooke, of Honolulu, one of our members, was reported. 

Enquiries were ordered to be made in England as to whether some early 
volumes of the ** Journal " could be reproduced by the Anastatic process. 



Digitized by 



Google 



HISTORY AND TRADITIONS OF RAROTONGA. 



By Te Akiki-taka-are. 



PART IV. 

(Continued froin Volume XXVIL, page 198. J 

[Everyone who lias studied the **Ilarotougaii Traditions'' iu their 
relation to those of the New Zealand Maori, must acknowledge the 
close connection between tliese two branches of the Polynesian race^ 
a connection which is further emphasized by tlieir having ancestors in 
common, as has been shown more than once in the pages of this 
** Journal." 

And yet it is somewhat surprising that the grand old epic of the 
New Zealand Maoris rehiting the birth of the gods as the offspring of 
the Sky-father (Rangi) and the Earth-mother (Papa), does not seem 
to have lield the same important place in the beliefs of the 
Rarotongans. This is somewhat difficult to account for, as it is now 
fairly well established that many of the New Zealand Maoris and 
Rarotongans formed pai-tof the second and probably largest migration 
into the Pacific from Indonesia, and wliich for convenience we use the 
Sanioan term ** Tonga-fiti," to distinguish it from the earliest 
migration (Samoan, Tongan, and some others), and from the third^ 
which, so far as can at present be ascertained, was of much later date, 
and, while including the East Coast of New Zealand Maoris, it also 
assisted in peopling the Hawaiian Islands and Tahiti, before it& 
members came on to New Zealand in tlie fourteenth century. 

This is a question that might well occupy the attention of the 
younger generation of Polynesian scholars now gradually coming to 
the fore, but it means a vast amount of study. There can be little 
doubt that the belief in the origin of all living things originating from 
the Sky-father and Earth-mother, was the primary belief of the 
Aryan- speaking people of India, dating probably from times antece- 
dent to their migration into India, that is, in the times of the 
Proto- Aryans, so called — for which see inter alia, Ragozin's **Vedic 
India," p. 136, and Dr. A. K. Newmau^s " Wlio are the Maoris?" 
p. 161, "Myth, Ritual and Religion," by A. Lang, and many other 
works. 



Digitized by 



Google 



56 JOURNAL OF THE POLYNESIAN SOCIETY. 

The question is too large a oue to be dealt with here ; but what 
follows may help to throw some light on it. In the 194 pages of 
oiosely written MSS. obtained in Rarotonga in 1897, lieing the 
teachings of Te Ariki-tara-are (said by Dr. Wyatt Gill to be the last 
of the high-priests of Rarotonga), the tradition given below is the 
only thing that approaches the Eangi and Papa story of the New 
Zealand Maoris, and as compared with the full detail in the records of 
the latter people, expressed in the teaching of the Maori College (see 
our. "Memoirs," Vol. III.)» it will be acknowledged to be a greatly 
attenuated account. 

First of all we note that the Maori name of the Sky-father (Rangi) 
is not used with the Rarotongans. Te Tumu appears to represent 
Bangi, whilst the Earth-mother is Papa in both accounts. Among the 
many meanings of tumu in Rarotonga, is that of ** the stem," **tlie 
foundation," whilst in Maori combination {tumuaki) it means ** the 
orown of the head," "the chief," and as tumu-rae a supreme chief or 
king. We may translate the word as " the original being," and is, in 
that sense, a god like Rangi. Papa (the Earth-mother) means " broad," 
"flat," as does Prithivi, the Sanskrit name of the Earth -mother. The 
Hawaiian name for the first man was Kumu-honua, in which Kimiu is 
their form of tumu — honua being the Maori whenua^ land. In this we 
Lave the same idea of " original," or " stem " as in Rarotongan. 

The Maori idea of the earth is that of a woman lying flat on her 
l)ack with her head to the east. 

It will be noticed that the Rarotongan Te Tumu is looked on as an 
ariki or chief — not as a god — and that the three eldest offspring of Te 
Tumu and Papa are quite unknown to the Maori account of the 
oreation of the gods. From Mr. Savage's note their names are seen to 
be emblematical terms for the accompaniments of child-birth. Te Uira 
(m, te uira) means lightning; Te Aa (m, 'Te Awha) means a tempest; 
Te Kinakina (same in m) means a full belly. After the birth of these 
three come three of the well-known gods, Tane, Tu, and Tangaroa^ 
oommon to the whole race. 

The next name (Te Nga-taito-ariki) is very peculiar in that it 
includes in it both the singular {te) and plural {nga). I do not 
remember a similar case in any dialect of the Polynesian language; 
and suggest that Nga-Taito-ariki was the name of a tribe or people, 
rather than that of one person — like Nga-Puhi of the Maoris. 

We are next told that certain priests of old, who appear to have 
flourished prior to the creation of man, searched the earth, and by 
*' knocking" on the earth discoveired man. This account seems some- 
what to be the Rarotongan rendering of the Maori account of the 
search of the gods for the female, eventuating in one being formed of 
earth, who, by the god Tane, became the progenitor of mankind. 



Digitized by 



Google 



Histary and Traditions of Rarotonga. 57 

From Te Nga-taito-ariki, the genealogy decends down to Tangiia- 
nui, the great ancestor of the Barotongans who flourished in the 
thirteenth century, 69 generations from the first named. But the 
account of liim will be found in a separate part to follow. 

A reference was made a few lines back to the belief of the Proto- 
Aryans in the origin of mankind from the Sky-fatlier (Dynus) and the 
Earth-mother (Prithivl). In Z. A. Ragozin's '' Vedic India," p. 160, 
we find tliis belief fully referred to, and he says, ** Numerous are the 
passages in wliich 'community of race' — kindred — is claimed with 
gods for men, explicitly, tliough in a general way : thus the verse, 
* We liave in common with you, gods ! the quality of brothers in 
the mother's bosom," is fully explained by this other: 'Heaven 
(Dyaus) is my father, who bore me ; my mother is this wide earth 
(PrithivT).' " 

Maori scholars will recognise in the last quotati(m a close 
approximation to a well-known form of Maori address. It was not 
etiquette among the Maoris to ask a stranger's name directly, but the 
following is the usual and polite way of ascertaining it. The one asks 
the other, ** Na wai taua?*^ (By whom are we two?) The reply is, 
*^ U/ Na Rangi mua ho Papa tana ! " (We two are descended from the 
Sky-father and the Earth-mother !) The first then says, " Nau mat 
tohi tuakana (m' taina) ! Kowai to taua tupuna? (Welcome my elder (or 
younger) brother ! who was our common ancestor ?) And from the 
answer to this the relationship becomes known. It will be seen that 
this is almost the exact equivalent of the quotation from the Rig Vida, 
or most ancient traditions of the Indian-Aryan people. 

The following is the original and the translation thereof from 
which the above notes have been made, the latter having been kindly 
supplied by Mr. S. Savage, of Rarotonga. It must be understood that 
this Pai-t IV. is the commencement of the history of Tangiia-nui, 
hence the leading paragraph.] 



A HISTORY OF TANGIIA-NUI. 



Translated by S. Savage. 

248. Tangiia, it is said, was a son of Pou-vananga-roa-ki-Iva, a 
son of Tupa and Oa-ariki, a son of Te Arutanga-nuku, a son of Te 
Maru-ariki, a son of Te Nga-taito, and a son of Te Tumu (i.e., he 
descended from those ancestors). (There was) a land named Avaiki- 
te-varing^-nui. An ariki (thereof) named Te Tumu, took as his 
wife Papa, and by her begat their three eldest children, namely, 
Te Uira, Te Aa, and Te Kinakina (see note for explanation), and 



Digitized by 



Google 



58 JOURNAL OF THE POLYHESiAN SOCIETY. 

afterwards came forth Kongo and Tane and l^uanuku and Tii and 
Tangaroa ; these were the godly sons. 

Then Te Nga-taito-ariki begat Atea, and Te Tupua-nui-o-Avaiki 
and Te Pupu and Ka'ukura, these were the arihi sons, the pro- 
generators of all ariki lines. (This sliould read : Then were begotten 
Te Nga-taito-ariki and Te Tupua-nui-o-Avaiki and Te Pupu and 
Ka'ukura ; these etc.) 

249. Now concerning those who opened up Papa (that is to say 
those who discovered the secrets contained within Papa the Earth- 
mother) they were named Rua-kana-kui, Eukana-kura, Ruakana-era 
and Kua-kaua-taunga, six all told ; their parent was Te Veka-o-te-po, 
alias Tongaiti, alias Tumu-ngao. They were the ones who opened up 
Papa. The following is the method or manner by which they did it, 
as was taught to us : — ^They came through the land called Avaiki-te- 
varovaro, otherwise called Raro-pu-enua, and, behold, they heard 
mumurings below ( within the eai'th). They listened, and from the 
sounds they heard came to tlie conclusion that the murmuring sounds 
came from a great number (of people). They said one to the other, 
^'Whence come those sounds?" Then said Rua-taunga, **Iknow, 
those murmurings come from within Papa." ** We shall see." Then 
some said, ** ! those people are great!" Another said, **What 
shall we do so that we may see ? " Then one exclaimed, ** Go slowly. 
Now let us move along slowly and pat the surface of Papa above and 
below, in the front and behind, inland and seaward, and thus we may 
discover all." After they had searched all about (that is all over 
Papa (the earth) ) they then commenced to knock on the earth, and 
made a gi-eat noise in doing so, and peered here and there intently 
until tliey came to a certain spot or place, and, behold, they saw from 
whence the sounds came. Hence they knew that they would eventually 
discover all the secrets contained within Papa. 

250. They then recited in thunderous voice as follows : — 

The Papa (earth) is growing up (asuming shape) 
*Tis the Papa of Avaiki — 
The Papa will stand up 
It will grow— and night is born. 
Then day is bom — 
Now it is heaving up — 
Now it is subsiding — 
Now we see Arii and Toro 
'Tis growing thick — 
'Tis growing solid 
etc. 



Digitized by 



Google 



History and Traditions of Rarotonga, 59 

251. They looked, and, liehold, what they did appeared to their 
minds to be good, and therefore they continued : — 

Papa (the earth) O ! the Papa will grow up 
The earth will arrive at maturity 
,, „ „ become ripe 
„ „ „ grow in height 
„ „ „ bud and flourish 
„ y, „ bear leaves 
,, ,, ,, become upright 
There will be many (people) in the east 
»» >> >> ?> >> west 

„ ,, ,, ,, ,, soutli 

„ „ ,, „ north 

252. The al>ove were the proceedings of the company of priests 
towards Papa (the earth) by which they became priests, and 
everything that should concern the descendants of Te Tumu and Papa 
became known to them and justified their titles of priests (iaunga). 

[Explanatory note rs portion of paragraph 248, which says: ** And 
by her begat their three eldest children, namely, Te Uira, Te A a, and 
Te Kinakina." 

More-taunga-o-te-tini tells me that the meaning of this is as 
follows, which I give in his own words : — 

" It is language that was used hy the priests in reciting the 
ancient legends to the people, but it really meant a description of a 
woman giving birth to a child. Te Uira was the shooting pains 
experienced. Te Aa was the pressing by the woman with her hands 
to her back and sides when the labour pains were severe and the 
delivery of the child was about to take place. Te Kinakina was the 
breaking foi-th of the fluid prior to the birth of the infant, and then 
were born Eongo, Tane, etc.," but he says these are only a few of the 
children mentioned, there were many others. 

Tangiia-nui — of whose history the above is the introductory part — 
was also a great high-priest. He taught in the Are-Vananga all the 
knowledge he obtained during his sojourn in Avaiki. His name as 
gi-eat high-priest was Te-ariki-tai-vananga-tara-keu-ki-te-rangi. 

More-taunga-o-te-tini (Tamuera Terei) is a direct descendant of 
Buakana-kura, one of the priests named in paragraph 249, on his 
father's side.— S.8.] 



Digitized by 



Google 



60 JOURNAL OF THE POLYHESiAM SOCIETY. 

E TUATUA NO TANGIIA-NUI * 

248. E tamaiti a Tangiia-nui na Pou-vananga-roa-ki-Iva, e 
tamaiti na Tupa-ma-oa-ariki ; e tamaiti na Te Ara-tanga-uuku ; e 
tamaiti ua Te Maru-aiiki ; e tamaiti na Te Nga-taito-ariki ; e tamaiti 
na Te Tumu. 

E enua ko Avaiki-te-vaiinga-nui ; e ariki ko Te Tumu. Te Tumu 
ko noo i tana vaine — i a Papa ; kua anau akera ana puke tamariki, 
tama mua tokotoru, ko Te Uira, ko Te Aa, ko Te Kinakina. 

Kua mama mai i muri i reira ko Kongo, ko Tane, ko Eua-nuku, 
ko Tu, ko Tangaroa. Ko te angai aitu la. 

Kua anau maira ko Te Nga-taito-ariki, ko Atea, ko Te Tupua-nui- 
o-Avaiki, ko Te Pupii, ko Kau-kura. Ko te angai ariki la. 

NO TE ARONGA I KANAIA I TE PAPA. 

249. Ko Rua-kana-kui, e Eua-kana-tea, ko Hua-kana-kura, ko 
Rua-kana-uri, ko Rua-kana-ero, ko Eua-taunga. Tokoono ratou; ko 
Te Veka-o-te-po to ratou metua — ko Tonga-iti rai ; koia a Tumu-ngaro. 
No ratou te papa i kana. Teia te tu e kite ei tatou e, na ratou i kana. 
Kua aere maira ra ratou, na roto i te kainga ra ko Avaiki-te-varovaro — 
koia a Raro-pu-eiiua. E ina ! e mumu teia i rare. Kua akarongo aere 
ua-o-rai ratou ; ** Teea ua teia mil ? ** Kua tuatua a Rua-taunga, **E, 
tei roto i a Papa ! '' Kua na-ko- maira tetai aronga e, ** ! e aronga 
tangata maata tena ! " Kua na-ko-mai tetai, " Kaakapeea tatou e kite 
ei ? " Kua manono maira e tokotai, na-ko-maira, **Eia, ka aere ana 
tatou, ka pakipaki aere ana i o Papa nei, i arunga, i araro, i a mua, i 
a muri, i a uta, i a tai, kia kite tatou." E oti akera ta ratou aaro 
aereanga i te Papa, kua topapa ratou i te papa i te pauuanga, e te 
maro, e te akeakeanga ; kia tae ra ratou ki tetai ngai, kua kitea, te 
tangi pauu ua ra i tetai ngai. Kua manako iora ratou e ka rauka. 

250. Kua rau-vanangananga iora ratou, ratou ua-o-rai, na-ko- 

akera : — 

Ka oi te Papa — 

Ko te papa ki Avaiki, 

Ko te Papa ka tu ki runga, 

Ka tupu, ka tumu te po, 

Ka tumu te ao, 

Ka eaau ki runga, 

Ka eaau ki raro, 

Ko Ari, ko Toro, 

Ka pararau-are, 

Ka patapata-tue, 

Ka matoru. 



* Expressed in the Rarotonga dialect. 



Digitized by 



Google 



History and Traditloni of Raroionga. 61 

251. Kua akara iora ratou, e iua! kua tau meitaki ta ratou 
akonoanga ; kua tia-ko-ak«ra ki te Papa : — 

' E Papa e ! ka iupu te Papa, 

Ka metua te Papa, 
Ka pakari te Papa, 
Ka rito te Papa, 
Ka roa te Papa, 
Kakao te Papa, 
Ka rau te Papa, 
Ka tia te Papa, 
Ka tini ki te itinga, 
Ka tiui ki te opunga, 
Ka tini ki Apatouga, 
Ka tiui ki Apatokerau. 

252. Ko te au angaanga tela a te arouga taunga i rave i a Papa, 
i taka ai ratou e, e arouga tauuga, ko te au mea te ka tupu i te uauga 
o TeTumu e Papa, i kitea puia ai ratou e, e taunga. 



PAKT V. 

DESCRIPTION OF THE FATHERLAND AND HISTORY OF THE 
TAMARUA FAMILY. 



Translated by S. Percy Smith. 

[The following, whilst professing to he tlie history of the Tamarua 
family of Harotonga, whose residence is at Nga-Tangiia, on the east 
coast of the island, really gives the most detail concerning the 
Fatherland of the Polynesians to be found anywhere in the traditions 
of the people. 

It is a strange sttny altogether, and describes how the gods 
assembled together with the chiefs and people to celebrate important 
functions. It was in this land of Atia-te-variiiga-nui, that the Sky- 
father, Te Tumu, and the Earth-mother, Papa, created the gods and 
mankind (see part IV. hereof), and to this country in the west the 
spirits of the dead departed to their final home, in the same manner 
as tlie New Zealand Maori says his spirits of the dead went to the 
Fatherland, Hawaiki-nui, or Irihia (which latter name Mr. Elsdon 
Be^t suggests, with great probability, to be tlie same ajs Vriliia, an 
ancient name for India, or some part of that country). 

In the book ** Hawaiki," it was suggested that tlie name Atia-te- 
variuga might be translated ** Atia-the-be-riced,"from vari^ a common 
name for rice in India and Indonesia ; and that meaning may still 



Digitized by 



Google 



62 JOURMAL OF THE POLYHESIAM SOCIETY, 

hold good ; but other meanings of varinga have since come to our 
knowledge. In ** Myths and Songs," Dr. W. Wyatt Gill, deriving 
Ids information from Mangaia Island, lying to the east of Harotonga, 
says that vari means "the beginning"; and in Barotonga itself 
varinga means ** the very beginning," or ** ancient." Either of these 
two latter names might be correct as implying the most ancient land 
the people were traditionally acquainted witli. That is, if vari and 
varinga are not derivative words from vari (rice), to wliich tlie meaning 
" ancient" has been affixed in later times, in the same manner that 
the word vavao with the Samoans now means ** ancient," though it 
probably was tlie name of one of the oldest countries that people were 
acquainted with, and which is retained in the traditions of otlier 
branches as that of an island. Dr. A. K. Newman traces varinga to 
tlie waringa or tcaringin tree, the sacred Ficiis religiosa of India.* 

The high-chief Tu-te-rangi-marama, who may be locked on as a 
king, is the great, ancestor of the Earotongans; and the building lie 
erected called Koro-tuatini, was the place of assembly of gods and 
men in the Fatherland. The word means the ** place of many 
enclosures," and being twelve fatlioms high could scarcely be built of 
anything but stone — it was no doubt a temple ; and the New Zealand 
Maoris, who have the same name, say it was identical with Te 
Whare-kura, the first and original house of learning at (or identical 
with) Te Hono-i-wairua, where the spirits of the dead gathered fioni 
the four quarters of the earth before departing, either to the supreme 
god lo in the twelfth heaven, or to Whiro, the evil spirit in Karo- 
henga — whicli we usually term Hades, though it was a very different 
place to the common acceptation of the meaning of that word. 

The history says, "It was also in this land that originated the 
wars that caused them to spread to all the islands," which statement 
agrees exactly with the New Zealand Maori account of the cause of 
the exodus from the Fatherland. 

At the fourteenth generation, or 350 years after the great chief 
Tu-te-rangi-marama, we come to Te Kura-a-moo, " who departed 
for the east and dwelt there " on account of troubles about some 
fishing matters. This appears to represent the New Zealand Maori 
account of the wars that originated through fisliing rights in 
Tawhiti-roa, which has been tentatively identified with Sumatra, and 
from other legends these ancestors of Tamarua probably went on to 
Java, which is one of their Avaikis, while one of the New Zealand 
Maori branches went to Tawhiti-nui, tentatively identified with 
Borneo. 

Sixteen generations (or 400 years or 750 yeai-s from Tu-te-rangi- 
marama) after Kui-a-a-moo, we come to Tamarua-metua of Avaiki 

♦ ** Who is the Maori ? " p. 222. 



Digitized by 



Google 



Hitioiy a/Ht Traditions of Ratoionga. 63 

(Java or Sumatra), and the statement is made that in his time '^ all 
Avaiki was broken up and scattered in small parties." This is a most 
important statement, for it seems to indicate— without stating the 
cause — the date at which this particular migratioki (the Tongafiti) 
became scattered and left Indonesia for the Pacific Islands, or perhaps 
moved on to some other parts of Indonesia, such as the Celebes, etc., 
and then oonuneneed the exploration of the Pacific, whether at once, 
or after a more or less prolonged further stay in Indonesia, we have no 
means at present of determining. 

For the use of future students it may be stated here liow tlie date 
of this exodus from Avaiki (Java or Sumatra) is fixed. A reference 
to our " Memoirs,*' Vol. IV., p. 12, in the note at bottom of page, it 
is shown how the period of Tu-te-rangi-marama was determined by 
the mean of four descents from him to the year 1900, and the date 
B.C. 475 arnved at. Taking the two descents mentioned previously 
in this paper (350 and 400), we get 750 years, which deducted from 
B.C. 475, gives the year A.D. 225 for this ** scattering of Avaiki." 

The table given below in the original Barotongan, from which these 
notes are compiled, does not agree at all with the four other 
Earotongan tables which correspond in themselves very fairly, and are 
therefore used in preference to the table of descent given below. This 
latter table has always seemed to the wnter to be open to question; 
firstly because it differs so wildly from all other known genealogies 
in length, and secondly because of the introduction of so many 
geographical names as those of ancestors — in tliis latter feature it is 
like the long Marquesan tables. It will be observed therein that there 
are fifteen Avaikis mentioned, and tlie question arises whether these 
are not really names of some of the Indonesian Islands, rather than 
ancestors' names ? 

In paragra[)h 525, we come across the celebrated navigator 
Ui-te-rangiora, whose name, however, does not enter into the 
genealogical tables of the Tainarua family; but it is to be found in 
the table at the end of the book ** Hawaiki," and there he is shown to 
be living fifty generations back from the year 1900. He was certainly 
one of the principal explorers and navigators of the race. For some 
account of his voyages see ** Hawaiki," p. 167 (third edition), and our 
"Memoira,*' Vol. IV., p. 37, where the discrepancy of dates between 
Maori and Barotongan accounts of this celebrated sailor are pointed 
out. According to the latter people he flourished about A.D. 650. 

It is upon such traditions as these tliat the whole of the chronology 
of the Polynesian race rests, and it is to be remarked that we shall in 
all probability never get anything better. The song^ included in the 
original text below it is impossible to translate as a whole without the 
aid of the Rarotongans themselves, and it is doubtful if they can do so 
at this time.] 



Digitized by 



Google 



64 JOURNAL OF THE POLYNESIAN SOCIETY. 

THE TAMABUA FAMILY OF ^ABOTONGA. 

[Translation.] 
51^, This is tbe history of Tamarua-te- Aia and Tamarua-te- 
akaariki and their families ; their ancestors, etc. 

He (the first named) was a descendant of Tamarna-mau-o-matatia, 
of Pouara, also of Tamarua-orometua, also of Tamarua-puangi, also of 
Tamarua-miti-rima, also of Tamarua-pai, also of Tu-terrangi-mamma, 
and also of Te Tumu. 

The (original) land (in which the latter lived) was Atia-te-varinga, 
and here was bom Te Tupua, son of Te Tumu : — 
Te Tumu 

Te Tupua-o-Avaiki 
Tu-te-rangi-marama. 

This man, Tu-te-rangi-marama, was the High Chief of Atia-te- 
varinga (the most ancient land known to the Barotougans, and 
identical with Hawaiki-nui and Irihia of the New Zealand Maoris), 
and he arranged for the constru.;tion of a most glorious sacred place 
(a building), and surrounded it with a great enclosure, very high, 
twelve fathoms in height; veiy spacious and of great widtli. It 
contained many beautiful things. It was a place where the spirits of 
mankind and of the gods assembled, as also those of the High Chiefs. 
It was called " The Koro-tuatini " of Tu-te-rangi-marama; and tlieu 
he appointed a guardian for the place, named lo-tini. (The Maons 
know of this name Koro-tuatini as a name for some temple in the 
^^ Fatherland " where the spirits of gods and men foregathered, but 
which is more generally called " Te Hono-i-wairua.") 

516. We shall see from what follows that that place was a 
gathering place of human spirits. The mother of Tau-toro — the son of 
Varo-kura — whose name was Varo-tea, when she died her spirit went 
to the place of Tu-te-rangi-marania, to the " Koro-tuatini." We also 
see it in that the spirit of Kuikui-tatau also went to that " Koro," 
and because it was an assembling place of all the spirits of men. It 
was the gathering place also of gods and the High Chiefs, because 
there originated the many wonderful tilings (such as customs, ritual, 
etc.) at Atia-te-varinga. We know it also through Tiki, who also 
went to that ** Koro " ; and through Ngaro-ariki-te-tara, who was 
carried there by Koura and Tiaka after the bathing in the dammed -up 
water of the sisters of Ngata-ariki, and they two handed her over to 
the Peneneki (water sprites)* who took her to Avaiki-atia. She had 
seen that Avaiki-atia. 

* Peneneki, acoording to Major Large, are water-sprites. But it is possible 
this name is synonymous with the New Zealand Turehu, Patn-paiarehe, or Pai-ehe, 
usually said to be fairies. They are light coloured, and from tbe description of 
them evidently a light-coloured race known to the Polynesians in the Fatherland. 
For the adventures of Ngaro-ariki-te-tara see Part III. of this history. 



Digitized by 



Google 



History and Traditions of Rarotonja. 65 

517. But the ptincipal and most delightful things at Atia-te- 
variiiga were the " Takmnias " (feasts, ceremonies, songs and dances) 
that were instituted by Tu-te-rangi^marama to animate (inspirit, 
glorify) his country ; such as the Pu (trumpet), the Pau (a dnini), the 
Kaara (a drum), the Eva-tipa, the Eiva-puapua-aki, the Eiva-tatau- 
maa, the Pen, the Eva-tapara, the Eva-tea, the Ura, the Akaao, the 
Karioi, and all the akaaka-puauau (? certain kinds of dances) at Avaiki. 

518. There was also the " Takarua-tapu " (the sacred ceremonies 
connected with the annual feasts of fii*st fruits), when there came the 
company of the gods to the assembly altogether at Avaiki — Kongo 
and his company, Tane, Buanuku, Tu, Tangaroa, Tongaiti, and 
Tu-tavake, with their separate attendants. There also came the 
High Chiefs, witli their attendants, at the command of Tu-te-rangi- 
marama of Avaiki. There came Te Nga-taito-ariki, Atea, Kaukura, 
Te Pupu, Te Uira, Te Aa, Hua-te-atouga, whose was the first part of 
the Takarua — named " likaa." This wtMB the oiigin of all the 
(succeeding) commands of old at Avaiki-atia. 

619 There were (assembled) there also the female gods: Taa-kura 
and her attendants; and Ari, Tupua-nui, Te Bangi-putai-ua ; Te 
Pao-o-te-rangi, and the daughter of Rua-te-atonga — Te Kura-akaipo — 
and tiieir respective attendants. 

520. There were also the Pereneki (water sprites),* very numerous, 
whose countiy was Te Rangi-topa-rere, whence they originated. 

Tu-te-rangi-marama, Te Tumu, and their attendants were also 
there. They were the (people who inaugerated) the institution of the 
"Takarua" at Avaiki. 

That was the land of splendid head-dresses of all the Chiefs, and it 
was from that land'of fnatea (? gorgeous plumes) that (these customs) 
spread to all the islands. It was also in this land that originated the 
wars that caused them to spread to all the islands. 

It was in this land that lived Te Tumu and Papa, and there also 
were born their sons and daughter. 

521. It was here that assembled the company of priests : Rua- 
kana-kiri, Rua-kana-tea, Rna-kana-kura, Eua-kaha-ero, Rua-kana- 
uri, Rua-taunga, and their attendants. It was the assembly of all to 
the one ceremony in that land, and it was they who selected the Chief, 
and they who set him up and annointed him ; that is, akaparapara. f 

(These things were done) to search out the correct procedure in the 
land, to search for means of safety and welfare for men, slaves, and 

* See above—probably a iniatake for Peneneki. 

t These names commencing with Rua, are probably represented in Maori by 
the names (also beginning with Boa) recorded by John White in *' Ancient History 
of the Maori," Vol. I., p. 3, but there said to be sisters of Tinirau^-sorely a 
mistranslation, judg^g by the meaning^ of the names. 



Digitized by 



Google 



66 JO^BHAL OF TH£ POLY/teSIAH SOCICTY. 

children. Such were the kind of men in thet land ; and it was these 
same people that spread over this great ocean. 
622. There was bom to Te Tupua 

Tu-te-raugi-niarama 

Bua-i-te-karii 

Te Ako'kara 

Te Ake 

Hoo-uri 

Moo -tea 

I 
Te-tiai-i-Avaiki 

Knra-a-moo 

Tau-tonu 

Amoa-aitu 

Te Ariki»ivi-roa 

Kake-toa-ariki 

Tua-ariki 

Te Ariki-okiotini 

I 
Te Kura-a-moo 

Te Kura-a^uioo departed for the east, and dwelt there, on account 
of trouble that grew up between him and his sisters: it was on account 
of a basket of fish-hooks which one of the sisters trod into the mud. 

523. He remained in these parts, and thei*e was bom to him : — 

Te Moo-take-kura 

(and then follows the names of 79 descendants (see the original) down 

to :— ) 

Tamarua-o-Avaiki 

Tamarua-nui 

Tamarua-paipai 

Tamarua-takinga 

Tamarua-taunga 

I 

Tamarua - oki -rua 

I 

Tamaraa-te-aranga 
Tamarua*pai 

He (the latter) lived in the times of Tangiia, and whom he joined 
(in the many voyages made by l^e latter, as the principal navigator). 



Digitized by 



Google 



Histoiy and Traditions of Hmrotonga. 67 

His home was at Papa-unri and at Ati-iuaono (on the south coast of 
Tahiti) and also at Pape-ete (on the north coast of Tahiti). At 
Mo'orea is a channel (Pin the reef), cut by Tamarua-pai, named 
Utu-kura. 

524. (Then follows 18 generations down to Tamaiua-orometua, 
who in 1897 was between 90 and 100 years old when I saw him at his 
home near Nga-taugiia, Barotonga.) 

525. In the times of Tamarua-nietua of Avaiki all Avaiki was 
broken up and scattered in $mall parties; everyone had become 
possessed of canoes. One eanoe was that of Ui-te-rangiora, and it was 
this one that enabled all Avaiki to scatter to different islands. In that 
same division (scattering) three of them joined, Tamarua, Te Aia, and 
Tai-vananga, and this was their (final) separation from (living in) a 
single land, Avaiki (or Atia-te-varinga), and they spread to every 
land from there. They (the three) embarked on the canoe of 
Ui-te-rangiora, and they it was who navigated tliat canoe to all the 
islands of the world. There was not any plaoe they did not visit ; and 
hence they became very expert sailors. When their vessel became 
rotten they renewed it, and when repaired they returned with their 
canoes to conquer Te Ravaki ; and when that was accomplished they 
went on to Bangi-ura and Mata-te-ra and subdued those islands ; at 
the latter place they killed the chief named Te Ango-noo-rangi, whose 
name was (aftei wards) given to a tapora (?mat) at the marae by 
Tamai'ua^ They also went to Nukare and subdued that land and all 
other islands, as fc^ows :— * 

The greater part (of the names) remain (unwritten); it was not 
possible to write them all. Those are the names of the islands that 
they visited. 

526. The following are the names of the lands in whicli they 
permanently dwelt in this ocean : Avaiki-ravo, i.e., Kuporu (Upolu) 
and Avaiki-ki-i*unga, i.e., Taiti and Iva (Tahiti and Marquesas) ; and 
tliey also went as far as Avaiki -tan tan, i.e.. New Zealand. Tamarua 
dwelt at Avaiki-raro, i.e., Samoa, but Avaiki was the principal name. 
He dwelt at Avaiki (Savaii), and Amoa was the name of his district,, 
where he had a fish-pond named '' Te Puapua-a-tiare." Tlie fish in it 
were the kanae^ that came in numbers in the proper season, from Yiti 
(Fiji). It was (? first) brought there from that land by Taniarua-te- 
ikumea, who had two sides (i.e. natures) ; one side was human, the 
other a fish. When he brought the fish he led tliem into the fish-pond 
by a channel from the sea ; a very small channel, probably like a 
ditch. By this channel the fish went to the pond, which was very 
large. Now, the men who had to watch it, went to the narrow place 
and there watched until the channel was full to the sides, and then 

* See list of islands in the original. 



Digitized by 



Google 



68 JOURNAL OF THE POLYHESIAH SOCIETY. 

they stopped up the channel in the narrow part near the sea, enclosed 
it closely, and then they stood on the banks and snared them, placing 
them in heaps, and carried them off. If the heaps {au) were not 
? consumed (the rest) were thrown away. 

Tamarua brought the name of that fish-pond with him to Raro- 
tonga, hence " Nga-te-tiare." 

527. (Sometimes) Tamarua dwelt at Taiti (Tahiti), at Pape-unri 
and Atimaono. Te A.ia and Tu-vananga lived at Pape-ete. At 
Moorea is the channel cut out by Pai (? Tamania-pai). (After living) 
at Tahiti (for a time) he would go to Avaiki-raro and stay there, 
(sometimes) to Autaria and stay there, and (even) to Atia-te-varinga. 
And so it was from generation to generation down to the time that 
Tangiia settled in Rarotonga, when these long voyages ended, even 
to the present time. 

528. In the times of A vaiki - te - paipai (? Tamarua -paipai) there 
grew up disputes at Avaiki-raro between Naea and his younger 
brothers. The origin of the quarrel was a spring orf water; one spring 
was set aside by the arihi for himself, not to be used by the younger 
brothers. A second cause was in connection with the division of the 
food as arranged by the an/:i, Naea, for himself; it was the youngei* 
brothers' business to arrange the distribution, the arikVs share and 
theirs. The kikomua (? fore -quarters) of the pigs and the ktkomua 
(? best parts) of all foods were for the artki alone, while the kik&niuri 
(? hind-quarters) and the (? inferior) parts of all foods were for the 
brethren. The younger brothers considered that they should take it 
turn and turn about, but the aviki would by no means consent to this. 
So the brethren seized the chief ; and thus commenced a new war in 
Avaiki, and its inhabitants were scattered far and wide. Naea fled to 
Avaiki-runga (usually means Tahiti and these parts, but here the Sage 
includes the Hawaiian Islands in that expression), that is, Vaii (ancient 
Tahitian and Maori name for Hawaii, i.e., Vaihi and Waihi) and 
Tangaungau (now called Lanai) at Vaii, and which are called " Avaiki- 
nui-o-Naea," and after he was driven out by the younger brethren. 
Avaiki belonged to Naea. 

529. Tlie names of those younger brothers were Tu-oteote, 
Karae-mura, Tu-natu, Kakao-tu, Kakao-rere, Uki, Pana, Pato and 
Ara-iti. 

(Then follow several songs). 

533. There were three men all of one family; Te Angai-aroa, 
was their father, Ati-aroko was the elder, Ati-aroa the younger, and 
Ati-kavera the youngest of all. They named the districts in which 
they lived (in Rarotonga) after themselves : that is, Aroko, Aroa, and 
Kavera. That man Te Angai^aroa was Tamarua (? which) and his 
children . 



Digitized by 



Google 



History and Traditions of Raroionga. 69 

NO TAHABUA. 

615. E tuatna no Tamarua-te-Aia-Pitimani, ma Tamarna-te- 
akaanki i to raua aiiauaiiga. 

E tamaiti iia Tamarua-*niau-o-inatatia i Pouara. E tamaiti na 
Tamarua-orometua i Aroko. E tamaiti iia Tamarua-puaiigi; e 
tamaiti na Tamarua-miti-rima ; e tamaiti na Tamarua-pai ; e tamaiti 
na Tu-te-vangi-marama, tamaiti a Te Tumu. 

E enua ko Atia-te-varinga ; e kua anau ta Te Tumu ko Te 
Tupua : — 

130. Te Tumu. 

Te Tupua-o-Avaiki. 

128. Tu-^-rangi-marama. 

Ko Tu-te-rangi-marama, koia te ariki o Atia-te-varinga; kua 
akataka aia i tetai ngai-tapu kaka, e kua akapini ki te koro maata e te 
teitei, okotai-ngauini ma rua te teitei. E koro atea maata, e te 
pararau-ar6, e ngai akakiia ki te au meitaki tini ; e ngai iiipaanga no 
te vaerua tangata, e uipaangano te au atua ravaitii ma te au tupuranga 
ariki ravarai. Ko te " Koro-tuatini " i o Tu-te-rangi-marama taua 
kainga ra ; e kua tuku aia i te tangata ki roto ei tiaki, ko lo-tini tona 
ingoa. 

516. Teia te niea i kitea e, e uipaauga vaerua taugata: Ko te 
nietua vaine o Tau-toru te tama a Varo-kura ; Yai'o-tea taua metua 
vaine ra ; ka mate ei taua vaiue ra, kua aere te vaei*ua ki o Tu-te- 
rangi-marama, ki roto i taua '^ Koro-tuatini" ra. Kua kitea katoaia 
ki te vaerua o Kuikui-tatau, kua aere rai ki roto i taua "Koro" ra. 
No te mea e ngai uipaauga no te au vaerua tangata purotu taua 
" Koro " ra. Kua uipa la te au atua e te au ariki ki reira ; no te mea 
ko te ngai ia i akatupuia4 te au ravenga katakata tini, ko Atia-te- 
varinga. Kua kitea ki a Tiki ; kua tae katoa aia ki roto i taua 
" Koro " ra. Kua kitea ki a Ngaro-ariki-te-tara, ka apaiia mai ei e 
Koura ma Tiaka i te paianga i te vai pa o nga tuaine o Ngata-ariki, 
tukuia atura e raua ki nga Peneneki, apaiia atura e nga Peneneki ki 
Avaiki-atia. Kua kite aia i tana Avaiki-atia ra. 

517. Teia te angaanga maata e te mataora i Atia-te-varinga, ko 
te takurua i akatupnia'i e Tu-te-rangi-marama ei akaau i tona enua ; 
ko te pu, ko te piiu, ko te kaara, ko te eva-tipa, ko te eiva-puapua-aki, 
ko te eiva-tutau-maa, ko te peu, ko te eva-tapara, ko te eva-tea, ko te 
ura, ko te akaao, ko te karioi, ko te au mea akaaka-puauau la i Avaiki. 

518. Ko te takurua tapu, koia te tere a te au atua ravarai i te 
uipaauga ki Avaiki i te ngai okotai ; ko Kongo ma tana tere; ko 
Tane nia tana tere ; ko Rua-nuku ma tana tere ; ko Tu ma tana tere ; 
ko Tangaroa ma tana tere ; ko Tonga-iti ma tana tere ; ko Tu-tavake 
ma tana tere : 

No te au ariki i to ratou uipaanga ki te takurua i Avaiki, i a o 



Digitized by 



Google 



70 JOURMAL OF THE POLYMtSUH SOCItTY. 

Tu-te-rangi-marama i te akonoanga i teianei takurua. Ko Te Nga- 
taito-aiiki ma tona pae ; ko Atea nia toiia pae ; ko Kau-knra uia tona 
pae ; ko Te Pupu ma tona pae; ko Te Uira ma tona pae; ko Te A a 
ma tona pae ; ko Raa>te-atonga nia tona pae ; nana te takurua 
mua — ko te li-kaa ; ko te akakapuaanga teia i te an akonoanga, i 
taito, i Avaiki-atia. 

619. No te au atua-vaine; ko Taa-kura ma tana tere; ko Ari 
ma tana tere ; ko Tupua-nui ma tana tere ; ko Te Rangi-pntai-ua ma 
tana tere ; ko Te Pao-o-te-rangi nia tana tere ; ko te tamaine a 
Rua-te-atonga — ^ko Te Kura-akaipo— ma tana tere. 

520. Ko te tere a Te Pereneke, manotini ratou ; ko Te Rangi- 
topa-rere to ratou niotia, ko te vaiinga la. 

Ko Tu-te-rangi-marama ma tana tere ; ko te ter» i a Te Tumu. 
Ko te au akonoanga teia no Araiki. 

Ko te enua pare-au tikai teia no te an ariki rnrarai, no teianei enaa 
matoa aere inai ei ki te pa enua katoatoa. E tupu rai i roto, i roto i taaa 
enua nei rai te tamaki i i*ato aere ei ki te pa enua katoatoa. 

Ko te Te Tumu e Papa, ko raua ana tei noo ki runga i taua kaiuga 
ra, ka anau ei ta raua tamariki tamaroa e te tamaine. 

521. Ko te kau taunga i ta ratou tere : Ko Rua-kana-kiri, ma tona 
pae ; ko Rua-kana-tea ma tona pae ; ko Rua-kana-kura ma tona pae ; 
ko Rua-kana-ero ma tona pae. Ko Rua-kana-uii ma tona pae ; ko 
Rua- taunga ma tona pae. Ko te uipaanga teia ki te akonoanga okotai^ 
ki roto i taua enua ra; e na ratou i iki te ariki, na ratou rai e kimi te 
ariki e, i urureiiga i te ariki — koia rai te akapai'apara. 

Ko te kimi i te tika ki runga i te enua, ko te kimi i te ora e ora ai 
te tangata, te unga, ma te potiki. Ko nga pae tangata teia ki roto i 
taua enua ra. E ko nga pae tangata i*ai teia i rato aere mai ei ki 
teia moan a maata nei. 

522. Anau ta Te Tupua, ko Tu-te-rangi-marama. 128. 



Moo-kura 

Rua-i-te karii 
125 Te Ake-kura 

Te Ake 

Moo-uri 

Moo-tea 

lo-tini-i-Avaiki 
120 Kura-a-moo 

Tau-tonu 

Amau-aitu 

Te Ariki-ivi-roa 

Kake-tua-ariki 
115 Tua-ariki 

Te Ariki-oki-tini 
113 Te Kura-o-moo 



Digitized by 



Google 



HMory and Tradiiions of Rarofonga, 



71 



Kua aere a Te Kura-o-moo ki te itinga o te ra, kiia noo atu aia ki 
reira, no te pekapeka i tapa ia raton ma iiga tnaine ; e kete mataa te 
ara, e taomi e tetai taaiiie ki raro i te vari. 

623. Kua noo atnra aia ki reira ; kua anau ta Kura-o-moo : — 



112 Te Moo-take-knra 

Avaiki-ia-atea 
110 Tu-veka-o-moo 

Te Tupua-me-neke 

Te Tupunga-aitu 

Te Pai-tua-tini . 

Avaiki-tia-inga 
105 Atia-a-mata 

Te Tupua-nui 

Te Tupua-iti 

Te Tupua-raunga 
100 Kau-tupua 
99 Te Einga-a-mata-o-Avaiki 

Tamarua-a-papa 

Tamarua-akaau-takurua 

Tamarua-metua-o-Avaiki 
95 Te Avaiki-po 

Po-atanga-o-Moo 

Vainga-atua-ki-Avaiki 

Te Uinga-ki-Avaiki 

Avaiki-tai 
90 Avaiki-na-oti 

Avaiki-miri-reka 

Avaiki-te-maora 

Avaiki-tupa-ero 

Avaiki-raita 
85 Avaiki-a-ora 

Avaiki-po- ta 

Avaiki-ka-rona 

Avaiki-mea-a-reka 

Avaiki-kokoia 
80 Avaiki-ka"-kite 

Avaiki-noe-aa 

Avaiki-toro-uka 

Te Angai-ariki 

Te Angai-atua 
75 Te Angai-aro 

Te Angai-taunga 

Te Angai-o-te-pereneki 

Te Angai-mua 

Te Angai-ki-roto 
70 Te Angai-ki-vao 

Te Angai-ki-a-tiki 

Te Angai-tapu 



Te Angai-tura 

Te Angai-rao-ta 
65 Te Angai-tiritiri 

Te Angai-enua 

Te Angai-ki-ura 

Te Angai-ariki-ki-Atia 

Te Angai-ki-tau 
60 Te Angai-moneka 

Te Angai-kl-Iti 

Te Angal-mauta 

Te Angai-tuma 

Mau-o-tai 

Mau-o-penea 
55 Mau-o-kapura 

Mau-o-kianga 

Mau-o-rangi-topa-rere 

Mau-o-to 

Mau-o-tere 

Mau-o-uru-o-kura 
60 Mau-o-rere-taua 

Mau-o-te-nere 

Mau-o-te-ringi 

Mau-o-te-niro 

Mau-o-te-angai 
46 Mau-o-te-riko 

Mau-o-te-eaau 

Te Itonga 

Te Akerua 

Nae-tu-kura 
40 Nae-iki-aitu 

Tua-ariki 

Taki-varu 

Te Pu-rara 

Avaiki-maine 
36 Tu-te*iku-mea 

Tu-mai-oni 

Tamarua-o-Avaiki 

Tamarua-nui 

Tamarua-paipai 
30 Tamarua-tukinga 

Tamarua-taunga 

Tamarua-oki-rua 

Tamarua-te-aranga 
26 Tamarua-pai (ko teia to 
Tangiia tuatau ia) 



Digitized by 



Google 



72 JOURNAL OF THE POLYMESiAN SOCiETY. 

Kua piri aia i reira ki a Tangiia; kua noo aia ki Pape-unri e 
Ati-inaono e noo atu ki Pape-ete. Tei Morea te ava tarai a Tamania- 
pai, ko Utu-kura te ingoa. 

624. Allan ta Taniania-pai : — 

Tamarua-keu 

Tamarua-viri-tai 

Tamarua-nui-rai 

Tamarua-tei-vao 

Tamarua-te-ariki, e-raka 

Tamarua-atero . 

Tamarua-tauraki 

Tamarua-tairi-i-te-rangi 

Tamarua-miti-rima 

Tamania-te-angai 

Tamarua-une 

Tamarua-pei-kuru 

Tamarua-puangi 
3 Te Mato 1 Tamarua-akua 2 Te Tauna-a-toka* 3 Te Mato 

Putanga-i-mate-kia-ruki Tamarua-tu-kaka , .4.^1.' 
Tumu-rakau Tamarua-kui-nau Tamarua-aere-marie 

Te LJa-takiri Tamarua-tavake 1 Tamarua-kaki-ta 2 Tua-ariki 

Tamarua-orometuaf Tamarua-mau-o-matatia 

Tamarua-te-aid-Piti-mana Tamarua-te-akaariki 

Tamarua-te.-ruaroa 

626. I to Taiiiarua-metua o Avaiki tuatau, kua pueu rikiriki a 
Avaiki ; kua rauka to pai no Avaiki katoatoa. to te pai no Ui-te- 
raiigiora, ko te pai la i pueu rikiriki ei a Avaiki ki te pa enua katoatoa; 
i taua pueu rikiriki-ke-anga ra, kua kapiti ratou tokotoni; ko 
Tamarua, ko Te Aia, ko Tai-vananga, ko to ratou topanga la ki vao 
me roto i te enua akotai, ko Atia-te-varinga, me te Avaiki katoatoa 
ki vao; e kia rato aere te tangata ki te pa-enua ravarai. Kua kake 
ratou ki runga i te pai o Ui-te-rangiora; ki a ratou te akaaere i taua 
pai ra ki te pa-enua katoa o te ao nei. Kare e ngai toe i te aere e 
ratou, e riro ratou ei aronga kite mauta i te akatere pai. E kia pe taua 
pai ra, ei reira ratou e anga aere ei i to ratou pai. Kia oti, aere ra to 
ratou au pai, e reira ratou e oki aere ei e ta aere ratou ki Te Ravaki. 
Kua ta i tei reira enua ki Rangi-raro, kua ta i tei reira enua ki Mata- 
te-ra, kua ta i tei reira enua; kua ta i te ariki, i a Te Ango-noo-te- 
rangi, kua riro mai taua ariki ra ei ingoa tapora na Tamarua i mua i 
te marae. Kua aere ki Nu-kare kua ta ; kua pera rai ki te pa-enua 
katoatoa : — 



* Te Tama-a-toka i Vai-i-kura. 

t Tamarua-orometua was living at Nga-Tangiia, Rarotonga, in 1897, and 
then between 90 and 100 years old. 



Digitized by 



Google 



History and Trailitions of Rarotonga. 



73 



Ki Nu-kare 
,, Nu-takoto 
„ Nu-taara 
„ Nu-mare 
„ Nu-pango 
„ Nu-it! 
„ Nu-tana 
„ Nu-ame 

Ki Iti-nui 
„ Iti-rai 
„ Iti-anaunau 
,, Iti*takai-kere 
„ Piti 
„ Pa-pua 

Ki Vaii 
„ Tavai 
„ Ngangai 
„ Maro-ai 



T 



► Probably the New Hebridies and Loyalty Islands 



The Fiji and Lau Groups. 



... Hawaii 



Tauai, now Kauai / ,, .. , , , 
, . > Hawaiian Islands. 



... Lanai 
... POahu 



Ki Tonga- nui 
„ Tonga-ake 
„ Tonga-piri-tia 
„ Tonga-manga 
„ Tonga -raro 

Ki Aval ki -raro 
„ Nu-taata 
„ Ma-reva 
M Pia 
„ Uea 
„ Raro-ata 
„ Amama 
„ Tuna 
„ Rangi-arara 
„ Rotu-ma 

Ki Vavau 
„ Niva-pou 
„ Atu-apai 
„ Tangi-te-pu 
H Rara 
„ Avaiki 
„ Kuporu 
„ Te Tuira 
„ Manuka 
„ Tokerau 

„ Uru-pukapuka-nui 
„ Uru-pukapuka-iti 



The Tonga Group. 

This name covers Samoa and the Fiji Group. 

Home Island 

Futuna 
Wallis Island. 

Northern Isle of Tonga Group. 
Great Hope Island 
Haabai, of Tonga Group 



Savai*i of Samoa Group 
Upolu „ H 

Tutuila „ t, 

Manu'a „ „ 

Union Group 
? Palmerston Island 



Digitized by 



Google 



74 



JOURNAL OF THE POLYHESIAM SOCIETY. 



Ki Enua-kura 
„ Iva-nui 
„ Iva-rai 
„ Iva-te-pukenga 
„ Te. Kirikiri 
„ Te Rauao 
„ Te Mae-a-tupa 
„ Rau-maika-nui 
„ Rau-maika-iti 
,, Ngana 

„ Te Pau-motu kotoatoa 
„ Akaau ... 



The Marquesas Group 



"The whole of the Paumotu Group" 
Fakahau Island 



Ki Taiti 
„ Morea 
„ Rangi-atea 
„ Uaine 
„ Taanga 
„ Porapora 



Tahiti 

Morea or Aimeo 

Ra'i-atea 

Huahine 

Taha'a 

Porapora 



The Society 

and 
Tahiti Groups 



Ki Rurutu 

„ Pa-pau 

„ Rima-tara, " Kati- 
pia ki Rima-tara, 
ka tipia ki Rauta- 
mea, e ra tonga 
koia e ki Mauke." 



[ The Austral Group 



Ki Mauke 
„ Motea-aro 
„ Atiu 
„ Auau 

„ Raro-tonga, oki atu 
ki runga ki 



Old name of Mangaia Island 



J 



. The Cook 
Group 



Ki Rapa-nui 
„ Rapa-iti 
„ Teni-te-ia 
„ Pa-pua 



Easter Island 

Rapa, or Oparo Island 



Ki Au-te-ria-nui 
,, Au-te-ria-iti 
„ Kateta-nui 
„ Kateta-iti 
„ Panipani-maata- 
one-okotai 

Te vai atu rai te nuiata, kare rava i ope te tataia; kote au pa-eiiua 
teia i aereia e ratou. 



Digitized by 



Google 



Hiiioiy um4 Tradftfons of Rarotwiga. 75 

626. Teia o ratou euua uoo taiuou i teia moawa iiei, ko Avaiki- 
raro, koia a Kuporu, e Avaiki ki riiiiga, ko Taiti, ko Iva ; e aere ua 
ratou ki Avaiki-tautau — koia **Nu-tiraiii." Kiia iioo a Tainarua ki 
Avaiki-raro— koia a Amoa; ko Avaiki te ingoa maata; kua iioo aia i 
Amoa, tona tapere, ko ^^Te puapua-a-tiare" te ingoa o tana roto>ika; 
e kanae te ika : ka tere mai ki roto, kia tae ki te tuatau, no Yiti mai 
tana ika ra, e ika tikina ki taua enua ra, ko Tamarua-te-ikumea te 
tangata i taoia mai ei taua ika ra. E rua tu o taua tangata ra, e 
tangata tetai pae, e ika tetai pae. Kua taoi maira aia i taua ika ra, e 
tae maira ki Ayaiki kua arataki aia no roto i te aya i tai, e aya oiti ua 
me te ayaaya ua nei paa te tu. Ka na reira taua ika A i te aerenga ki 
uta i te roto ; e roto maata taua roto ra, ina ra, ko nga tangata i te 
akara, ka aere la ratou ki tai i te ngai oiti i te nia i taua ayaaya ra, ka 
akara ua rai ratou e ma te ki takiri ua te roto e pari ki te pae ; te arai 
ratou i tai i taua oiti ra, te koro ra, e pin tikai e, ei reira roa*i e 
tutuia e te tangata, kia taei aere, ma te au aere, ka tauta ua. E kare 
e rauka i te au, akaruke atu. Kua taoi mai aia— a Tamarua — i te 
ingoa o taua roto-ika ra ki Rarotonga nei — koia a " Nga te Tiare." 

627. Kua noo a Tamarua ki Taiti ; ko Pape-uriri e Ati-maono to 
ratou nooanga. Ko Te Aia e Tai-yananga ka noo ki Papa-ete. Tei 
Morea te aya tanii a Pai, (Tamarua -pai ?) Ka noo ki Taiti ka aere 
ki Ayaiki -raro ka noo ki reira ; ka liere ki Au-taria ka noo i reira ; ka 
aere ki Atia-te-yannga. Ka pera ua ra i tera uki, i tera uki e tae ua 
mai ki te tuatau e aere mai ei a Tangiia ki Raro- tonga nei, ko te 
mutunga mai nei la, e teia noaM. 

628. I te tuatau no Ayaiki -te-paipai (? ko Tamarua-paipai la paa ?) 
kua tuputupu te pekapeka i Ayaiki-raro, ko Naea e nga teina. Tera te 
tumu i te pekapeka i a ratou, ko te puna-yai ; okotai ua puna rai i 
akatakaia e te ariki nona anako, auraka o te ai teina. Tera te rua ; ko 
te tuanga kai i akonoia e te ariki, e Naea nana, tei i a ratou katoa oki, 
tei te ai teina, te akaaereanga i te tua i te kai i te koutu- ariki, e rua ua 
yaenga i te tuanga i te kai, ko ta te ariki, ko ta ratou. Ko te kiko-mua 
i te puaka ma te kiko-mua o te au kai rayarai na te ariki anake la. Ko 
te kiko-miri i te puaka e te kiko-miri o te au kai rayarai na ratou ia. 
Ko te manako o te au teina, takitai i runga, takitai i raro. i^are roa 
te ariki i tuku mai ki ta te au teina ; opukia io ei e te aronga teina ; 
akatupuia te tamaki ou i a Avaiki, ko Avaiki ia, e te pueu-rikiriki ke 
ra. Te oro ra a Naea ki Avaiki runga — koia a Vail. Ko Tangaungau 
i Va*u (?) ra, i tuatua mai ei e, " Avaiki-nui-o-Naea," e nga teina, e nga 
teina oriori marie ; no Naea rai a Avaiki. 

629. Tera te ingoa o taua ai teina aue kai ra : Ko Tu-oteote, ko 
Karae-mura, ko Tiori, ko Tu-natu, ko Kakao-tu, ko Kakao-rere, ko 
Uki, ko Pana, ko Pato, ko Ara-iti. 



Digitized by 



Google 



76 JOURNAL OF THE POLYNESIAN SOCiETY. 

630. E AKATABA NO TAMARUA-TAIRI-TE-RANGI 

1 . E ! e toro, e tnron turou, e aitu, 
E tupuranga taua i Avaiki, 

Ka riri ngana, ka tot<> vero o tera, 

E karere tei aere mai nei« 

Ko Marere, ko Rua-a-toa, 

Maroto i te ranga rau, no Are-munamuna 

I te rapitoanga ara a Tangfiia 

Takn ariki e, ko mei Avama ra, 

E oo ka tupu ana. 

2. E ! e t<»ro e, i tupu i Avarua 

Te tupurauga ariki, te ariki tapu aite 
Me roto i te kaponga a Rangi-o- Atea 
No Arai-te-tonga, te pu i te tamaki, 
I aaia ai te raui tangata o Tangiia 
Me runga me Anga-ta-kura, 
£ iki mataiapo tei vao ariki o Taiigpiia, 
Ko Tairi anake te karakia atu, 
Ki nga atua, kia tapu roa ra, 
E oo ka roa, rire e. 

3. Tena ka ui, ka ui, ka ui au ana, ra toro e, 
Te ui Akimere e torn e, 

Ko Toa oki, ko Pare ko Tea, 

Urua ae e taku tama • 

Tera tupuranga toa mei taito 

Kaore e korero, ko Tairi anake. 

Te toa rongo nui o Tangiia 

E uira, e rapa, e aruru i te rangi. 

Te karonga tamaki a Tairi 

Naringa, e taku tama ! 

E takiri tu i te aereng^, 

E iku, e poroaki ei kinei koe, 

Ka ano au ra e. 

4. Tena te rongo, to rongo i te matangpi, 
Matangi, matangfi i au ana ra toro e, 
Te rongo i te matangi torn e. 

Ko te rangfi pua noa te ae ki raoka 

Ko Papa, ko £i-tara, ko Tautu-tapuae, 

Mokora ; ko te toa i aere i te uru o te kare, 

Ko Tairi anake te toa rongo nui o Tangpiia 

E uira, e rapa, e aruru i te raugpi 

Te karonga tamaki a Tairi 

NHringpa, e taku tama ! 

E takiri tu i te aerengpa, 

B iku, e poroaki ei kinei koe ^ 

Ka ano au ra e, oo ka roa rire e. 

5. Tena oatu, oatu ra te rongo, o te rongo, 
Te rongo i au ana ra toro e, 

Oatu ra te rongo ki Make -vero, torn e, 



Digitized by 



Google 



HMory and Traditions of Raroionga, 77 

Ki Arai-te-tonga, ki taoa tamaine— 

Ki a Tea-amru-o-te-rangi 

Kua euumiumi nOa a Te li-o-te-ra-tanoku, 

E tama e ! e tei ea Tairi ka ngaro oi nei, 

Kua riro ake nei ki te ugaki kaiuga 

Akairi matakeinanga no Makea, 

Anraka e vaiio i Arai-te-tonga . 

Ka amutia a rairingao, 

Ko Tairi anake te toa rongo nui o Tangpiia, 

E uira, e rapa, e aruru i te rangi, 

Te karouga tamaki a Tairi, 

Naringa, e taku tama ! 

E tukiri tu i te aerenga 

E iki, e poroaki, ei kinei koe, 

Ka ano au ra e— e — 



Tena ka ranga, tuetuea tei ipo ua 

Ai a Kamana e au ana ra toro e, 

Kua karo tapere te tamaki a Tairi, 

Tei uta te rotopu te ngaru rapa, 

I tua te koko ki Ati-ua 

Kua koi Tairi i te rau-pupa>a-toa 

Ei apaipai, ei kare puapuaaki ki Iti-kau, 

Ko Tairi anake te toa rongo nui o Tangiia, 

E uira, e rapa, e aruru i te rangi, 

Te karonga tamaki a Tairi 

Nariuga, e taku tama ! 

E takiri tu i te aerenga 

E iku, e poroaki ei kunei koe, 

Ka ano au ra e— e — 



I'ena, teia, teia te mea kino, ka kinokino, 

Ka kino i au ana, ra toro e, 

Teia te mea kino e, 

E ekai taeake na raua io 

Ei kau te nio, te ki pepeke, 

Ko nga tapere e torn, 

Tukua ki Ro-riki-ina-ve ma Tokerau, 

Nga pokoinui e rua i tangarua ia 

A Akaoa ma Are-renga 

I rauka mai oi, akairi ake oi, 

Ki Arai-te-tonga, ki te vaka e nui, 

Koia Taki-tumu, 

Ko Tairi anake te toa rongo nui o Tangiia 

E uira, e rapa, e aruru i te raiigi, 

Te karonga tamaki a Tairi, 

Naringa, e taku tama ! 

E takiri tu i te aerenga, 

E iku, e poroaki, ei kunei koe, 

Ka ano au ra e — 



Digitized by 



Google 



78 JOUBHAL OF THE POLYKESIAM SOCHTY. 

8. Tena ka inoe, ka moe, ka moe i an ana ra toro e, 
Ka moe te angai no Ngati-Tangiia 
Tei mua, motia ei Iti-kanapa, 
Te vairanga ko fce rotopii nei, 
Ei tara tamaki te vairatiga, 
Oatu tei miri, motia ei rau-tiutiu, 
Kua rere te toa rango nui ko Tairi, 
Ki runga ki Tara-rau, tu eru atu ei, 
Akareki atu ki Vai-moko, 
Tukua ki Tuaunga, e taiiia e Tairi ! 
Akapou to oro ki runga Mannga-roa, 
Akaruke atu oi, 

Ko Tairi anake te toa rongo nui o Tangiia 
E uira, e rapa, e aruru i te rangi, 
Te karonga tamaki a Tairf, 
Naringa, e taku tama ! 
E takiri tu i te aerenga, 
E iku, e poroaki ei kuuei koe 
Ka ano au ra e, e oo ka roa rire e. 

[Tera atu te au pee te vai nei, kare e tataia i kuuei ; ko 

** Te pee i te akatara o Tamarua-tairi-te-rangi.** 
** E pee no Tamarua-koroia,'* nona te tuatua ra, 
' E aa i oro ana Takinuku ' ; i mate ki te tamaki.] 

533. Ko te au ingoa tela no 4ie anan a Tamarua, me te mua e tae 
ua atu ki te ope : — 

Ko Te Ei-tuputupu Tamarua 
,, Tamarua-Te-Aia-Pitimana-Te Ariki-na-vao 
,, Putiki-aumea-ki-atu-Tamarua 
,, Tamatea-Iakopa-Atapa-Tamarua, 
,, Te Ua-takiri-arapaki-te-Ariki-maro-Tamarua. 
,, Noia-tooa-tauira-aki-Te Angai-Tamarua. 



Digitized by 



Google 



THE LAND OF TARA 
AND THEY WHO SETTLED IT. 



THE 8T0BY OF THE OCCUPATION OF TE WHANGA-NUI-A-TAKA 

(the great harbour OF TARA) or port NICHOLSON, 

BY THE MAORI. 



By Elsdon Best. 



PART VII. 

(Coutimt$d from page 17, Vol, XXVIII.J 
THE MAORI TEXT. 

It has been considered advisable to preserve in print the Maori 
Text on whicli is based most of the story of "The Land of Tara." 
Although this text will interest but few of our members, it is 
nevertheless advisable in the interests of liistorical and philological 
research tliat the original sliould be printed for fear of accidents to 
the MS. volumes in whicli it is contained. 

The original volumes are in possession of Te Whatahoro (the 
Scribe of our " Memoirs," Vols. III. and IV.), but copies are in 
private hands, and from these latter the text now printed below has 
been taken. The matter itself is the result of some years' work by 
the Scribe, who wrote it out to the dictation of the old priest of the 
Whare-wanaiiga, Te Matorohanga, who had been taught in his youth 
in that Maori college the ancient lore of his tribe. As the whole is 
expressed in the purest Maori language as spoken by the people 
prior to the advent of the European, it may serve to counteract the 
tendency to a bastard style of Maori now becoming prevalent in the 
writings of the present generation. 

TE WHANGA-NUI-A-TARA. 

Ka haere mai a Ngai-Tara, ka noho ki Motu-kairangi 
me era atu wahi. 

NA, i te whitu o Maehe i te tau 1867, i a matau i Kete-pakaru, ka 
mea mai a Te Waitere, a Kereopa, ki a Moihi Te Matorohanga : — 
" E koro ! Korero ra ki a matau ko wai nana i noho mai nga tahataha 
tai nei, timata mai i Heretaunga uei liaere mai nei ki Wairarapa nei.'' 



Digitized by 



Google 



80 JOURNAL OF THE POLYNESIAN SOCIETY. 

Ka mea atu a Moilii Te Matorohauga : — " E ta ma ! Kua liolia au 
i te korerotanga i iiga ]corero tauawhi a nga taiigata pupuri o te 
wananga ki a koutou, he kotikoti rahi uo koutou i aku korero iua 
korero au." 

Ka mea atu a Kereopa:— '^E ta! Ko tou papa auo, ko Te 
Ura-o-te-rangi, e whakararuraru ana i au korero." 

Ka roea mai a Te Matot-ohanga : — " E pai anahei korero noa ake 
ma tatau. Me timata mai e au te whakatakitaki mai ki a koutou i 
Turanga-nui-a-Rua, i te rawliiti roa." 



I te wa i welie mai ai a Wliatonga i a Toi-te-huatalii i Whakatane, 
ka mea a Whatonga ki te tipuua, ki a Toi : — ** Hei konei noho ai i to 
kainga, kia haere ake au ki te whakatakitaki haere i o mokopuna ki 
nga wahi o te whenua nei e watea, ana i te noho a te tangata, tirotiro 
haere ai he taunga iho mo te remu ki raro, a maku e hoki mai ki te 
toro mai i a koe i konei." 

Ka mea atu a Toi : — ** Haere i te taha rawhiti o te motu nei, kei te 
watea te whenua, ka kite taunga koe hei te talia moana e koe te 
taunga tangata, kia rua ai nga rourou, ko to te moana, ko to uta, hei 
matua mo te pani me te wahine, me nga tamariki. Haere, ko to 
matua iwi waiho kia tuwhera ana, hei ara kuia, hei ara mo nga 
tamariki." 

Ka whakaae atu a Whatonga ki a Toi mo ana tohutohu mai ki a ia. 

Ka haere mai a Whatonga me ona tamai'iki, me ton a ope, e tata 
ana ki te rua rau tane me nga wahine, haunga nga tamariki. Ka tae 
mai a Whatonga ki Ohiwa, tae mai ki Huiarua, ka mahia tona kainga 
ki reira. Ka roa e noho ana i Huiarua, ka neke mai ki tetahi wahi 
ano noho ai ; ka mahia tona wliare, a ka oti, he wheki nga pakitara 
me te tuarongo, ko te whatitoka he rakau. Ka kawaia taua whare ko 
Tapere-nui-o- Whatonga te ingoa, ka mutu. 

Ka roa e noho ana, ka neke mai a Wliatonga me ona tamariki, me 
tona ope katoa, noho rawa mai ko Maraetalia. Kaore a reira i tau ki 
a ia, ka neke mai i Maraetaha tau rawa mai ko Nukutaurua. Ka roa 
e noho ana i kona, katahi a Whatonga ka korero ki a Tara raua ko 
Tautoki (Enei tamariki na Hotu-waipara, wahine a Whatonga, nana 
a Tara, i moe ra i a Umu-roimata, nana a Wakanui, a Tiwhana-a- 
rangi. Na Reretua, wahine tuarua a Whatonga, nana a Tautoki- 
ihunui-a- Whatonga, na Tautoki a Rangitane-nui. Ko te tuahine o 
Tautoki ko Rere-ki-taiari. Kati i konei) — ka mea atu ra a Whatonga 
ki a Tara, ki a Tautoki : — ** E ta ma ! Haere ki te mataki i te whenua 
e takoto nei. Kia mama he hoa mo korua, waiho ake nga wahine kia 
pepeke ai korua. Kei te rongo korua ki nga kupu mai a to korua 
tipuna i poroporoaki mai ra ki a tatau." Ka mea atu a Tara:— '*Ae, 
e marama ana au ki ona kupu katoa." 



Digitized by 



Google 



The Land of Tata. 81 

Ka whiriwhiria te ope a Tara me tona taina, e toru tekau topu 
taua ope a raua. Ka baere mai, tae mai ki Te Wairoa, haere tonu, 
tae rawa ai;ii ki Heretaunga, ka noho i reira mataki haere ai. Ka 
haere mai, tae mai ki Raiigi-wliakaoma, Iiaere tonu mai, ka tae mai 
ki Okorewa, kaore i tau ; ka liaere, ka tae ki Para-ngarehu, katahi ka 
titiro. Ka mea atu a Tara ki te taina * Ka tau tenei wahi/ Ka tae 
ki Porinia, ka haere tonu, ka tae ki Rangi-tikei ka ahu ma roto i te 
awa, ka tae ki Patea, ka maro ki Tongariro, tae atu ki Taupo. Haere 
tonu atu ma uta, ka tae ki Titi-o-kura, ka tapatu ki Mohaka, ka ahu 
ki Te Wairoa, ka tae ki Nukutaurua. Ko Taka-raroa te kainga i 
noho ai a Wliatonga me ona tamariki, me ona mokopuna, me oiia 
tangata. 

Ka tae atu a Tara ma me tona ope, ka nui te koa mai o Wliatonga 
kua tae mai ona tamariki. E tata ana ki te tau e ngaro ana taua ope 
a Tara raua ko te taina. Katahi ka korerotia e Tara me te taina, me 
ta raua ope katoa te ahua o te takoto o te whenua, te ahua o nga 
awa nunui e puta ana ki roto ki nga moana wai tai, ki nga moana 
wai maori. Me te ahua o nga moana e uru ana te wai tai, me nga 
moana e uru te wai maori, ara moana wai maori tuturu. Me te 
ahua o te takoto o nga maunga, me te ahua o nga ngahere, me te ahua 
o nga mania. Pau katoa te korero i te ahua o nga taha moana me te 
ahua o te takoto o te whenua o nga tahataha moana, pau katoa te 
whakamarama. 

Ka ui atu a Whatonga ki a Tara : — ** Ki ta korua titiro kei 
tewhea wahi te matua o te tanga* e ora ai ? " 

Ka mea atu a Talra me Tautoki : — ** Kei te pongaihu tonu o te 
motu nei, kei reira nga motu e rua i rongo ra tatau i tapaia e Kupe 
ki nga ingoa o ana tamahine, ki a Matiu, ki a Makaro, Na, ko te 
motu nui rawa kei te pu o te tonga, kei te puau o nga rerenga e rua 
ki waho ki tahora nui a Hine-moana ; engari ma te uma tangata tenei 
e noho. Engari nga motu ririki e rua, ma te waka anake e whaka- 
taetae. Ahua pai te motu ki te taha mauru, tera e tipu te ora o te 
kopu ; ko te motu i te talia rawhiti (Ward Island) he horehore, he 
kopuru te one." 

Ka ui ano a Whatonga : — **E pewhea ana te motu rahi e ki na 
koe mo te kumara? " 

Ka mea atu a Tara: — ** He pai te oneone, he onematua, he 
tairanga te ahua o nga akaaka, ehara i te mahora, he one rere te 
wai." 

Ka mea atu a Whatonga : — ** He one maroke ma te tau hauwai 
e tipu ai te kai." 

Ka mea atu a Tara: — "Ae, etahi wahi ahua maru ana, e pai 
ana." 

* Possiblj in error for tangata. 



Digitized by 



Google 



82 JOURNAL OF THE POLYNESIAN SOCIETY. 

Ka mea a Whatonga : — *' Tena, lie hohoiiu te waLaapu o ngaawa 
na ki ta korua titiro, me ka timu te tai ? " 

Ka mea atu a Tara : — ** Tetahi, te wahaapu o te tab a mai ki te 
rawhiti. Ko te ngwtuawa ki te taha maiiru he pae onepu, kei te pae 
moana o te taha moana tutuki rawa ki te moana o te tuawhenna. Ki 
taku whakaaro ake tera pea e whakaikeike te ngutuawa.*' 

Ka ui alio a Whatonga : — " Kaore he raiiga kowhatu o waho ? " 
Ka mea a Tava : — " Kaore, kei nga ripa rawa nga kowhatu e 
whakaika ana." 

Ka mea and a Tara: — **He tata mai tera motu ki te titiro atu, 
tera pea e po tahi ki te moana ka u atu ki uta ki te whakaaro ake.'' 

Ka mea ano a Tara : — " Ko te moana o te taha mauru (Porirua) 
he wai tai, he moana pai, he moana ruru i te hau. Na nga pae hiwi 
i wahi te hau ki ta matau titiro; he onematua te one. He pai te 
ngutuawa o taua moana, engaii he uaua ki te noho takitahi, ma te 
umauma tangata anake e noho. Kotahi te^ motu kei waho o te 
ngutuawa e takoto mai ana ; tera pea e tae te waka ki te hoe atu i te 
atatu kia poutumaro te ra ka eke ai, ki te whakaaro. He motu pai 
tera ki te titiro atu, he maru, he papa tiraha, he kainga hau te ahua, 
engari e pai ana hei matua mo te waliine me nga tamariki. Na, te 
moana wai maori (Wairarapa lake) o te taha rawhiti o te pae 
maunga, he tahora nga tahataha, he moana kai ki te whakaaro ake, 
he repo nga tahataha, e heke ana nga awaawa o te pae maunga. He 
pae maunga kei te taha mauru, kei te taha rawhiti ; he ahua papaku 
nga pae hiwi o te taha rawhiti. Ko te taha mauru he pae maunga 
ranga kowhatu, one takataka, one parahuliu, he tohetea te one-, lie 
tauranga pualieiri, e hara i te liuka matua. Ko taua pae maunga he 
pokohiwi no te motu nei e takoto tonu mai ana, a tae mai ki te moana 
kei te taha mauru o tatau e noho atu nei. E heke ana nga awaawa o 
nga pae maunga e rua, o te taha mauru, o te taha rawhiti, ki roto. 
Ko te awa whakaheke o te moana nei kaore i rahi, e heke ana ki te 
talia maraiigai ki ta matau titiro. Kotahi rawa te motu paku nei e 
tau mai ana i waho i te taha rawhiti. He one punga nga tapa o taua 
moana, ma te umauma tangata ano e noho tenei wahi e taea ai. Pera 
hoki ra te moana i korero ake ra au i te pongaihu o te motu nei, taha 
rawhiti o te pae maunga ra, ma te umauma tangata e noho. He uaua 
te whakamatua ki te titiro noa atu, engari he pai te oneone, he one 
kai, he one paraumu, he one matua, he one pakirikiri etahi wahi ; he 
pai nga mania, he tuwhera. 

** Na, kotahi o nga moana ina tonu e tuwhera mai ra, he moana pai 
ano tera, he moana wai tai, e rere ana mai nga awa o te tuawhenua 
ki roto, engari he umauma tangata ano mana e whakatau ki te noho 
a reira. Kati, kua taunaha au i te moana wai tai o te pongaihu o te 
uiotu nei, ko reira he okiokinga mo tatau." 



Digitized by 



Google 



The Land of Tara. 83 

Ka mea a Wliatanga : — ** E pai ana. Kaua hei haohao nui te 
noho a kotitoii i te whenua i haere na koutou. Ka pa kua ururuatia 
te noLo a te tangata e liaohao imi ai te whakaaro nobo i te whenua ; 
ko teuei, whakatatanga ta koutou haere ki te whakanoho i te moana 
o te poiigaihu o te motu iiei, te moana wai tai, te moana wai maori." 

Ka whakaae nga tamariki, katahi ka wliakatika te ope o 
Whatoiiga ratau ko oiia tamariki. Ko etahi o oua tangata i wai]io e 
ia ki Nukutaurua noho ai, hei tiaki i tuua wahi ; e noho tahora noa 
iho ana te noho a te tangata. 

Ka haere a Whatouga, a Tara, a Tautoki, me era atu, nga tane, 
nga wahine, me a ratau tamariki aiio. Ka tae ki Heretaunga, ka 
mihi a Whatonga ki te takoto pai o te whenua. Ka tae ki Raiigi- 
whakaoma, ka mea ia kia u o ratau waka ki reira nolio ai, kia ta te 
ugenge o nga tangata. Ko Akaaka-nui te wa i heke mai ai te ope o 
Whatonga kawe mai i on a tamariki ki Poneke nei, tae atu ki 
Wairarapa nei. 

Ka noho ra i Bangi-whakaoma ra, a ka roa, ka pae lie kai ma 
ratau, he ika, he aruhe, ka haere ano te ope o Whatonga, ka tae ki 
Okorewa. Ka haere tetahi o nga waka ki uta o te moana o 
Wairarapa mataki haere ai, ka mutu, ka hoki mai ; ka haere ano te 
ope nei, ka tae ki Poneke, ka u nga waka ki te motu e kiia ra ko 
Matiu, ka noho i reira. Ka tahuri ki te mahi whare, he mea tiki nga 
rakau i uta, me nga rau whare. Ka oti nga wliare, ka tahuri ki te 
mahi papa whakatipuranga kai knmara, korau, hei kao. Ka oti enei 
mahi katoa, pae rawa te kai ki roto i nga rua, ara te kumara, ko te 
korau he mea mahi ki runga i te wliata karaho. 

Ko te whare i mahia ki reira mo ona Imnaonga, ka oti te mahi, ka 
kawaia aua whare e torn, ko Haere-moana tetaJii, ko Aotearoa tetahi, 
ko Te Pu-o-te-tonga tetahi. Enei whare he whakanoho, ara he tino 
wliare, engan he hopara makaurangi te whakairo o aua wliare e toru. 
Ko aua ingoa mo to ratau liaerenga mai i te moana i Hawaiki ki te 
kimi haere mai i a Toi, a Haere-moana. Ko te motu i noho ai ratau, 
i mahue atu ai Hawaiki, te kainga tuturu ; a Aotearoa, ara, na Kupe 
tera ingoa i tapa Aotearoa, mo te haerenga mai i te moana, kaore i 
kite whenua, a tae mai ana ki tenei motu, koia a Aotearoa. Na, ko 
Te Pu-o-te-tonga, ko tenei ingoa mo te mahuetanga o ona tamariki, o 
Tara, o Tautoki, me to raua tuahine, a Rere-ki-taiari, m^ ona 
mokopuna, ki te pito rawa ki te tonga o tenei motu noho ai, hei 
whakamaharatanga mo te wahi i a ia. 

Ka kotahi te ngahuru e noho talii ana a Whatonga me ton a 
whanau, me ona mokopuna, a Tuhoto-anki, a Turia, a Hine-one, a 
Eangitane-nui, me etahi atu o ratau, me te iwi hoki i weliea mai hei 
awhina i ona tamariki. Kotahi rau topu nga tane, nga wahine, ka 
wehea e Whatonga, kotahi rau takitahi ki a Tara, kotahi rau takitahi 
ki a Tautoki-ihunui-a- Whatonga. Ko te tuahine o Tautoki, ko 



Digitized by 



Google 



84 JOURNAL OF THE POLYNESiAM SOCIETY. 

Rere-ki-taiari, i mauria e Whatonga ki Nukutaurua hei tiaki i a ia. 
I tena wa kua kaumatuatia a Whatonga, kua mokopuna tuariia. 

I muri o te otinga o euei wliare, ka liaere a Whatonga me Tara 
me etalii tangata hoe waka kia kite i Te Mana o Kupe. Ka tae ki 
Matakitaki, ka hoe atu ki Kapiti, a ka hoki mai, ka tae mai ano ki 
Matiu noho ai. Ka haere ki te mataki i nga ngntuawa o te moana, 
me te motu nui o waenganui o aua awa e rua ; ka hoki mai ki 
Matiu. 

Ka tae ki te Ihouui, ka ki atu a Whatonga ki ona tamariki tane 
me o raua iwi e rua ra : — ** I muri i a au ka hoki nei kaore a konei e 
tau hei kainga tuturu. Waiho tenei kainga mo nga wahine me nga 
tamariki ; haere nga tane ki roto i te motu wawahi rakau ai, kume- 
kunie aka, hei mahi vvhare, Iiei niahi pa. Me mahi te pa tuwatawata 
ki te wahi i poua ra e au taku pou, kia pau taua wahi ki roto i te 
tuwatawata. Kia tu te pa, ko te puna wai hoki kia tata e koutou ; 
me whakaara ano te tuawata (?) ara haerenga ki te wai, pekerangi ra a 
waho, koi tutakina e te taua whakaeke, kia taea ai he wai. Ko nga 
whata kai hei runga i te tihi o te puke ra. Ka hanga i te tuwatawata 
ki te wahi i poupoua haeretia ra e tatau, kia rahi tonu. Kia kotahi 
tonu te waha ngutu o te ara nui, ka tuwatawata ai te ara nui, ka 
hanga ai nga puhara, kia rua ki te tomokanga, kia rua ki te urunga 
ki roto i te marae nui o te pa, kia toru nga puhara mo tetahi taha, 
mo tetahi taha. Me take nunui katoa nga take, kia rua wana liei 
whakauru i waenganui o nga take, kia kore ai e taea e te taua. 

Ko nga wahi manuka, rarauhe, hurupi, me tahutahu katoa ki te 
ahi i nga tau katoa e kitea atu ai kua ururuatia, koi mauria mai e te 
taua hei whakapuru i te take o te tuwatawata ka tahu ai ki te ahi, 
koinei te take e peratia ai e koutou nga wahi ururua. 

Kia toru nga tuwetawata, kia kotahi te tuwatawata matahao, kia 
kotahi te tuwatawata pekerangi, ka tae ai ki te tuwatawata matua, ko 
to roto rawa tena o nga tuwatawata.'* 

Ka mea atu ia i penei ai tana tohutohu kia noho pai ai nga 
kaumatua, me nga wahine, me nga tamariki. 

Ka mea ano a Whatonga : — ** Na, kia mahara, he pa koraha tenei 
no koutou, he pa mate i te kai kore, ko te rakau tena ma te taua, he 
whakaawhi kia mate ai i te kai, ka horo te pa. Me hanga e koutou 
he whare tatara ma koutou. Kia maha tonu ki nga pakitara tuarongo 
o o koutou whare tu ai, hei whatanga ika, hei pukaitanga kao 
kumara, korau, hei karaho punga pipi, kuku, paua maroke ma 
koutou. Mo te puta rawa ake he taua, e tu ana te poutaka o te kai, 
o te ai-uhe, o te karaka, o te tawa, te ora o to koutou tipuna, o Toi- 
kai-rakau,* e kiia nei ko Toi -kai -rakau." 

♦ ? Toi-te-huatahi. 



Digitized by 



Google 



The Land of Tana. 85 

Ka mea ano a Wliatonga : — *^ Nga papa mara e tata mai ana ki te 
pa nei, e pai ana, ka moe lioki nga tangata ki roto i te pa nei. Engari 
nga mara i matara atu ki tawhiti, me hanga rawa he pakokori tiaki 
paenga mara. Erua ona take o tenei tu pa, he tiaki i nga papa mara 
kai, he whakaaraara i te pa tuwatawata, ka mohiotia he taua ka 
huaki te taua." 

Ka mea ano a Whatonga ki ona tamariki : — " Me wehe te pa 
tuwatawata o tetahi o korua ki te puke i te taha katau o te ngutuawa 
i te taha rawhiti ra ; kia penei ano te mahi o te pa. Engari ko nga 
tino kaupapa mahinga kai hei te motu nui i kiia ake ra. Ko nga 
whare tatara kia pera ano me taku i korero ra te ahua o te mahi, o te 
tu hoki. I peratia ai kia titiro ai te taua whakaeke kaore he whata, 
he pataka takotoranga kai, ka mohio e kore e roa e awhi ana kua 
mate tenei pa i te kai kore, ka noho tonu ki te awhi i te pa. Na, ka 
mohio koutou ka roa e awhi ana, ma te mate kai e tute kia haere. 
Mehemea e taea e koutou te whakaeke i tetahi wahi ke noa atu, e 
kore e roa ka hinga tena taua, kua mate hoki ratau i te kai kore. 
Koia te take i waiho ai nga kai i roto i te whare tatara. I penei ai 
au ki a korua kia wailio ai tetahi o o korua pa hei whakahauora i 
tetahi, ka paaha te taua ki tetahi o o korua pa, ma tetahi e tiki 
mai e whakauru, a ka pera hoki ki tetahi." 

Ka mea ano a Whatonga: — "Na, tetahi mahi ma korua, me 
haere ki te titiro i tetahi wahi pai e kore e kitea e te tangata haere, 
era e uaua te kimi a te tangata, ka mahi he kainga ki reira. Ka 
kawe he kai takoto roa ki reira, he ika maroke, he kete aruhe paka, 
he paua maroke, he taha huahua, he karaka tao maroke, he kete tawa 
tao whakamaroke. Ka kiia tena tu kainga he kainga punanga, kia 
kore ai e kitea. Hei te po an ake e tahu ai te ahi kai, kanaka i te 
awatea, koi kitea te auahi e koiri ana. Tenei punanga mo nga pa 
horo ; ka whakamau ki reira nga morehu noho ai, a mo te rongo 
ranei he taua kei te haere mai, e wliakahoro ana nga wahiue, nga 
kaumatua, nga tamariki ki reira noho ai. Kia watea te pa tu tahora 
mo te toa anake, kia mataratara te tu a te tangata, kia kore ai lie 
raruraru ki nga koroua, ki nga wahine, ki nga tamariki. Ki te oti i a 
koutou enei tohutohu aku, e kore koutou e rarua i te taua." 

Ka mea atu a Tautuki ki a Whatonga : — ** He pai rawa a runga i 
nga moutere nei mo nga kaumatua me nga wahine me nga tamariki 
hei nohoanga." 

Ka mea atu a Whatonga : — ** Kaore, ko te he o tena ka mohio te 
taua kei reira nga koroua, nga wahine nie nga tamariki, k^ waiho te 
nuinga o te taua kia awhi ana i te pa, ka tikina aua moutere ka 
hunhia e te taua." 

Ka mutu nga tohutohu mo te ahua o te maiii pa, me nga punanga, 
me nga pakokori, me nga whare tatara, ka mea a Whatonga: — "E 
torn nga rakau hei ako ma koutou, he tokotoko, he rakau poto, he 



Digitized by 



Google 



86 JOURMAL OF THE POLYMESiAK SOCiETY. 

taialia, he pouwlienua, kotahi tonu enei e rua. Kaua liei noho noa iho 
ki te ako i te hapai o enei rakau. Ko te tino tohu mau nio enei rakau 
kei nga timu o te tangata ; ka kite koe e oi ana nga timii pakihiwi, 
kua mea ia ki te patu i a koe. Ko tetahi o nga tohu, rae titiro e koe 
ki nga konui o te waewae wliangai kaua te waewae tarewa, hei te 
waewae whangai anake koe titiro ai, hei te tinm hapai o 
te rakau. Mehemea e rua nga ringa e pupuii ana i te toko- 
toko, i te taiaha, i te pouwhenua ranei, me titiro koe ko tewliea te 
ringa taki o te rakau. E rua hoki ringa o te tangata matau, he katau 
tetahi, he maui tetahi ; kia pera ano tau hapai i tau. Ki te whiti ia 
ki te maui, me whiti ano e koe ki te maui hoki tau, a ki te whiti ki 
te ringa katau, me whiti ano tau ki te ringa katau tau rakau mau ai, 
me te u ano o mata ki tona timu, ki tona konui ranei. Kaua enei 
toliutohu e taka i a koe, kia kite ai koe i te patu mou. 

Na, mo te rakau poto kaore he tikanga. Mehemea ki te mau koe 
i tau ki to ringa katau, pai noa atu ; a ki te mau mai ia i tana ki tona 
ringa maui, pai noa atu — ^he rakau poto lioki tena. E rite ana tena ki 
to ringa tonu ake nei. Na, ki te whakapiri korua ko to hoariri, kia 
taka o waewae, kaua hei tu, whakataka ana ia kia hangai tona aroaro 
ki to kaokao, whakataka ana hoki koe kia liangai tonu to aroaro ki 
tona aroaro. Engari kia tere tonu to whakapiri ki to hoariri, ka 
whangai e koe kia wawe ia te turapa. Kaua e tukua te mata o to 
rakau kia tarewa ki waho, wailio tonu i mua i to waewae whangai te 
mata o te tokotoko, o te taiaha, o te pouwhenua ranei. Ma te mata e 
whakaoho, ma te rapa e patu mehemea he taiaha, he pouwhenua 
ranei. Mehemea he tokotoko, kia tata tonu te mata o to tokotoko ki 
to waewae whangai, taki ki tetalii kinga a waha. Ko te patu potOj 
liei raro te rapa o to rakau poto, tetahi hapai tera. Ko tetahi liapai, 
wailio kia tarewa ana to ringa i waho, kia wawe ia te whawhaia mai 
ki te tuku mai i tana patu ki roto i a koe. Ka kite koe kua tuku mai 
tona patu ki roto i a koe, me tuku ma to ringa maui e poka to karo, 
kia watea ai to ringa i te patu ki te whiu i te patu, kia hoki rawa ake 
te ringa i tona patu kua pa ia i to patu. Te wahi tere e mate ai te 
tangata ko te pu o te taringa, he rahirahi te angaanga o kona, e kore 
e hoki te tuarua o te patu ; kaua te tinana, a ki te pai te watea o te 
ringa hapai rakau, topea e koe i te ringa, hei raro iho i te timu o te 
pakihiwi tope ai, engari kia tarewa te ringa ki runga ka tope ai.'* 

Ka mea a Whatonga : — " Mo te whakaputa ki waho o te pa ki te 
hoariri, kia tere te puta o nga toa, kia tokorua te rerenga kotahi, 
pena tonu a luuri mai, a muri mai. Ko nga toa ki mua kaua hei 
puta ka tu mai i waho, oma rawa ki te wahi watea, waiho ma te toa 
e whai, kia watea ai te waha ngutu ka tahi. Nga take o te pei*a kia 
whai ai nga toa ki te mataika mana hei takiri i tona ingoa toa ; ka 
watea a muri i nga toa. Mehemea e kapi ana a waho o te ngutu, 



Digitized by 



Google 



Th^ Land of Vara. 87 

kokiiitia tonutia ki waenganui toiiu o te matua, engari kia iiiaki tonu 
te rere o nga toa liapai rakau ki mua. E kore teiia ope e tu i a 
koutou, ahakoa pewliea te nui o te ope taua." 

Ka mea aiio a Whatotiga ki ona tamariki : — '^ Meheniea he ope 
taua tau, tukuna kia tekau nga taiigata kakama te waewae ki mua, 
kia ahua matara atu ka tuku atu ai i nga (kaikape). Mehemea 
kotahi tekau nga toa hapai rakau, kia tokorima e tuku i muri tata 
tonu mai o nga tangata waewae, ara kia ahua matara atu ano. Ki 
te whakaeke nga toa o tera taua, kia tata tonu mai ki nga tekau 
tuatahi ka manukawhaki ai e to tangata waewae. Ki te liuri mai to 
aroaro o te ttekau tangata tuatahi, kia ata haere atu te tokorima toa 
taua, me te karanga atu — * Tahuri ! Tahuri ! ' hei whakatau noa, o 
kore e taliuri atu. Me tuku mai ki muri o te tokorima toa taua taua 
tekau, ka tahuri atu liei whare mo nga toa tokorima ra. Na, ko nga 
toa taua e rima i wehea ra mo muri me noho i muri tata tonu o te 
tekau tangata waewae ra. Ki te pena te whakahaere, e kore e taea. 
Ko te tiiio matua ki muri mai aki atu ai ; kia kaha te whakatitiko 
haere." 

Koia teuei nga tohutohu a Wliatonga ki a Tara raua ko tona 
taina ko Tautoki-iliunui-a-Whatonga. Ka mutu nga tohutohn, ka 
akoua e Whatonga te karakia mata rakau, me te karakia lioa i te 
tapuwae whai i te tangata. Ka mutu ena mahi a Whatonga, ka ki 
atu a Whatonga : — " Ko te mauri o te tino pa kawea e korua ki te 
taha ki raro o te paepae o te turuma o te pa, ki reira takoto ai. Hei 
te kowhatu huka-a-tai, hei te kowhatu onewa ranei, kaua etahi atu 
kowliatu: ka whakanoho ai ko Tuhinapo, ko Tunui-o-te-ika, kati 
kia.rua, koia na nga atua noho paepae whakaheke o mua iho; ko 
Maru tetalii atua pera. Euei atua hei tiaki i te pa, hei whakaatu kei 
te haere mai te taua, hei whakaatu i te tohu mate o te taua, o te pa 
ranei.'* 

I muri o enei tohutohu ka mea a Whatonga mana e haere, mana 
e hoki mai ki te toro mai i a ratau. ** Kia ngaro to koutou tipuna ki 
te hopara nui a Papa-tuanuku, hei reira whakanekeneke mai ai i o 
korua taina me te iwi ki tenei taha o Te Wairoa whakanoho haere 
mai ai, a ma koutou hoki e whakanoho haere ake tenei pito o te tai 
rawhiti o te motu nei. Ki te tae mai he heke i muri i a koutou tukua 
atu ki te taha hauauru noho ai, kia mau a Te Mana o Kupe, o Kapiti, 
i a koutou, hei okiokinga koroua, wahine, tamariki ma koutou i nga 
wa e haere ai koutou i te ope taua. Kaua nga motu nei e waiho hei 
okiokinga koroua, wahine, tamariki ranei, koi wehewehe i nga toa 
taua hapai rakau. (Mo Matiu, mo Makaro enei kupu a te kaumatua 
ra). Engari kia kaha te whakangau i nga tangata kakama te waewae 
te hapai rakau, kia waitaka ai te whakaputa rakau i te aroaro tangata 
ke." 

Ka hoki a Whatonga i konei. 



Digitized by 



Google 



88 JOURMAL OF THE POLYNESIAN SOCIETY, 

I muri i a Whatonga, i te niarauia o Putoki-imi-o-ta«, ka timata 
te wbanau te auga ki te whakaepa rakau, aka-tokai, mo nga whare, 
mo te pa lioki i poua haeretia ra to ratan papa, e Wliatoiiga. 

Ka patai i kouei a Kereopa: — "E Moi ! He aha te tau o iiga mabi 
e korero iia koe, kia marama ai tatau ki iiga tau o te noboanga 
baeretanga mai o te motu iiei tae mai ki iiaiaiiei ?" 

Ka niea a Moibi : — " Kaore be taubere o 11 ga tau i te iwi Maori 
penei me ta te Pakeba uei. Ka umtu ano uga mea i marauia ko iiga 
marama me iiga ra, me te raumati, me te botoke." 

Ka mea aiio a Te Matorobaliga : — " Me boki ano taku korero ki a 
Tara raua ko te taiua me o raua iwi. Ka mabi rakau ; ko etabi ki te 
mokibi rakau, ko etabi ki te tuku rakau ki raro; ko etabi ki te 
tuporoporo, ko etabi ki te wawabi, ko etabi ki te tarai tata, ko etabi 
ki te amoamo ki te taba o te awa e kiia uei ko Heretaunga, katabi ka 
mokibi ai ki rawabi o Te Wbauga uui a Tara. 

Engari kia marama auo, i te wa i a Wbatonga i webe mai ra i 
ton a tipuna i a Toi, kaore ano a Obiwa i tapaia, a Hinaioia. a 
Turanga, a Maraetaba, a Nukutaurua, tae mai uei ki Te Wbauga 
uui a Tara, a Porirua, a Kapiti, engari a Matiu, a Makaro, a Te 
Maua o Kupe ; kua tapaia era wabi e Kupe ratau ko oua tamariki ; 
uga wabi i uobo ai, i baere ai ratau, i wbai iugoa. Na Katoraugi, be 
mokopuna tuatoru ua Kupe raua ko Hiue-te-aparangi taua tangata e 
ki ana, na tona tipuna i korero ki a ia uga iugoa o uga wabi i tapaia 
baeretia e ratau ko on a tamariki, o te motu uei. 

Kati, ka wbakaarabia te pa uui touu, ara te pa tuwatawata, ka oti, 
ka niabia uga wbare o roto. E rua nga tino wbare wbakanobo, ko 
Haukawa tetabi, lie ingoa na ratau mo te moana i waenganui o tenei 
motu, o tera motu ; ko tenei wbare no Tautoki. Ko tetabi o nga 
wbare wbakanobo ko Wbare-rangi, be wbakaraaliara mo te wabi i tu 
ai a Wbarekura i Te Hono-i-wairua, i Uru. Ka tapaia te ingoa o te 
puna wai ko Te Puna o Tinirau. Tera ingoa o ratau mo te bauga e bu 
na te pakake i te moana, koia na Te Puna a Tinirau. Ka tapaia e 
ratau te iugoa o te pa ko Te Wlietu-kairangi. Ko te take o tera ingoa 
kaore e kite tangata o etabi iwi ke, ka uobo moke uoa ibo ratau, ko 
nga wbetu anake o te rangi nga mea bei uiatakitaki ma ratau i ia po, i 
ia po. koia a Te Wbetu-kai -rangi. 

Ka oti te pa nei katabi a Te Umu-roimata ka ki atu ki a Tara : — 
^* Ko te moana nei me waibo te ingoa ko koe." Ka wbakaae atu a 
Tara, koia Te Wbauga uui a Tara. 

Ka mea atu a Te Umu-roiuiata ki a Tara : — " Me whakaara be pa 
kia toru ki tera taba ki te tuawbenua bei mataki taua baere mai, ope 
baere mai rauei, o nga iwi o tawbiti, kia owbiti ai koe a Te Wbetu- 
kairangi. Kia tu era pa liei wbakaruru i Te Wbetu -kairangi, 
Koi tibaebaetia matau te tangata i te ra e wbiti ana." 



Digitized by 



Google 



The Land of Tara. 89 

Ka whakaae a Tara kia mahia aua pa e toru. Ka mahia a Uruhaii 
ki te pito ki te tonga, ka oti tena. Ko Te Maioha te whare iiui i roto ; 
ehara i te whare whakanoho tenei whare. 

Ka whakaarahia ko Te Aka-tarewa, he pa ano, kei te taha tonga o 
Matairangi (Mt. Victoria) te tunga o tenei pa. Ko te whare o roto ko 
Moe-ahuru, e hara i te whare whakanoho. Ka oti tenei pa ka 
whakaarahia ki te inataniata o taua taukahiwi e hangai atu ana ki te 
marangai. Ka ara tena pa ka tapaia te ingoa ko Te Wai-hirere; te 
take o tera ingoa mo te mate makutanga o nga wahine me nga tane i 
tetahi marangai nui, tu ana te wai i ruuga i taua kiirae i te nui o te 
ua, he mea kari ki te awa-kari, katahi ka rere te wai ki te moana ; 
ka tapaia he ingoa mo taua pa ko Te Wai-hirere. Ko te whare o roto 
ko Waipuna, he ingoa no te wai o taua pa, he puna kari te wai, ka 
waiho hei ingoa mo te. whare. 

Na, ko te walii uaua o ta ratau malii ko te mahinga niai i nga 
rakau. Ko te roroa o nga pou matua o Te Whetu-kairangi me te pa o 
Tautoki, a Para-ngarehu, i te rae o te taha rawhiti ra te tunga o Para- 
ngarehu, he pa nui hoki tera, engari kaore i rite ki a Te Whetu- 
kairangi te nui. Kati, ko te roroa o nga pou matua e toru whanganga 
te roroa ; ko nga wana e rua whanganga. Kotahi te hau ki te 
whenua o nga pou matua me nga wana. Ko nga huapae o te pa e 
wha, he mea hohou ki te aka-tokai katoa. Ko nga pou tal^e o te pa e 
rinia whanganga te roroa; ko te nunui o aua pou, o te take kotahi, 
kotahi whanganga. Ko nga pou matua kotahi te hau te matotoru, 
haunga nga wana whakauru penei ano me nga wana pa o naianei 
engari ko te roa e rua whanganga me te hau. Ko te hohonu ki roto i 
te oneone e kotahi te hau. Na, ka kite uiai koutou i te nui o te mahi 
me te tainiaha o aua rakau, me te whakaterenga atu i te wahi i mahia 
ai tae atu ki tera taha o Te Whanga nui a Tara, mo aua pa, tae atu ki 
nga pa pakupaku, me o rataU whare ano. Ko te kainga punanga 
nohoanga mo nga wahine me nga koroua me nga tamariki, me ka horo 
i nga pa tahuri, i nga parekura ranei, ka kawea ki Takapau-rangi kei 
te kauru o Wainui-o-mata, he moana kei te taha rawhiti o Te 
Wiianga-nui-a-Tara, kei uta atu o te pa o Para-ngareliu te pahT 
punanga. 

Na, he maha nga papa mahinga kai ; ko Kirikiri-tatangi kei te taha 
rawhiti o Te Whetu-kairangi. Ko Maraenui kei te taha mai ano ki 
Te Au a Tane nei. Ko Huri whenua ko te walii e kiia ra ko Te Aro 
tae noa mai ki raro iho o Tawatawa i te taha mauru marangai. Ko te 
wahi i waiho ra hei papa takaro ra ko Hauwai tera maara. Ko te 
wahi e kiia ra ko Watiwhaania (Watt's Farm, Newtown), taua 
ta whanga ra puta atu i te taha mauru o Uruhau ra, ko Pae-kawakawa 
tera wahi, he papa kumara na Hine-kiri. He wahine rangatira tenei, 
he uri na Tara, he tuahine no Wakanui ; ko Hine-kiri to mua, ko 
Wakanui to muri, ko Tiwhana-a-rangi to muri iho; na Hine-akau 



Digitized by 



Google 



90 JOURNAL OF THE POLYKESIAM SOCIETY. 

anake ratan, waliiiie ariki a Tara, lie mokopuna a Hine-akaii iia 
Whata. 

Ko tera pa ko Te Wai-liirere no Te Raugi-kai-kore, he tama teiiei 
na Tuhoto-ariki, tuakana o Turia ; he taiigata raiigatii^a teiiei, pai, 
manaaki tangata ia. I man mai tetahi wahine me ana tamariki 
tokotoi-u, no Mua-upoko, i te taua a Ngati-Rangi. Ka tae uiai ki 
Uruhau-, i a Pakau taua pa a Uruhau, ka tukuna taua wahine e 
Whirikai, rangatira o Ngati-Rangi, ki a Pakau, hei utu mo te kete 
manga me te kete aruhe i haere ai ratau i te taua ki a Mua-upoko i te 
matenga i Pukehou ; he pa tera kei te rawhiti o Otaki, a Pukehou. I 
mau mai taua wahine i reira, me ona tamariki, ka mauiia oratia mai 
hei utu mo aua kai. I reira a Te Rangi-kai-kore e noho ana i te wa i 
tae mai ai a Ngati-Rangi. Ka tukua taua wahine me ona tamariki ki 
a Pakau, ka mutu, ka whakatika a Pakau, ka mea mai ki a Te 
Rangi-kai-kore: — " Kia rua ki a koe hei kiuaki kumara mau, kia a-ua 
ki au hei kinaki kai ma taku tamahine ma Whakapiriuha." 

Ka mea atu a Te Rangi-kai-kore ki a Pakau raua ko Whirikai : — 
^* E tama ! Kia torn rawa matenga mo te tangata ? Ko te pa horo, a ko 
te tukutuku hei utu kai ma korua, a ko te ki hoki kia patua hei 
tamiwaha ma korua ? E hara tenei i te whakaaro tika na korua." 

Katahi ia ka karanga atu ki taua wahine, ki a Hinerau : — " E 
liine ! Whakatika ki runga, hoake tatau ko o tanaariki kirotoi"Te 
Whetu- kai rang], ki te okiokinga o te tangata." 

Ka mauria taua wahine me ana tamariki tokotoru; ka tae ki reira, 
ka ki atu a Wakanui : — " E ta ! Te Rangi-kai-kore, haere, kawea te 
wahine ra me ona tamariki ki Pukehou, ki to ratau na kainga noho ai. 
Ka tika tau, kia torn rawa matenga mo te tangata i te ra tahi; ka ora 
ano ra, kia ora, kaua e whakamokaikaitia te tangata." 

Koia au i ki ai he tangata rangatira pai a Te Rangi-kai-kore. 



TE TAUA A MUA-UPOKO KI TE WHANGA NUI A TARA ; 
KA PAAHA KI TE WHETU-KAIRANGI. 

Na, taua pakanga a Ngati-Rangi nei kei te take i haere ai a Te 
Kopara, rangatira o Mua-upoko, ki Patea ki te amio i a Rauru, i a 
Ngati-Ruanui hei ngaki i te mate o Mua-upoko i Te Pukehou, i mate ai 
te rangatira o taua pa. Ka haere mai nei a Tamatea-kopiri, a 
Kakataia, nga rangatira tenei o taua ope taua a Te Kopara. 

Kaore te taua nei i tika atn ki roto o te awa o Heretaunga ki Te 
Hau-karetu, ki Pa-whakataka, ki Parihoro, ki nga wahi i noho ai a 
Ngati-Rangi. Ka tika ke mai taua taua ki Hataitai, ki Uruhau, ki 
Te Aka-tarewa, ki Te Wai-hirere, kia tahuri enei pa, kia taea ai e 
ratau a Te Whetu-kairangi, te tino pa ariki o te motu ijei o Motu- 
kairangi. Na Hine-kiri tenei iugoa i tapa, be kianga atu na Tara ki 



Digitized by 



Google 



Th^ Land of Tara. 91 

a Te Umu-roimata: '* Wai ra he ingoa mo to tatau nei motu e noho 
nei tatau?" Ka mea atu a Te Umn-roimato, "He iti hoki a Te 
Whetii-kairangi ? " 

Ka mea a Tara : — ** E, no te pa ano ra tena ingoa." Ka karanga 
atu a Hine-kiri : — ** Waiho be ingoa ko Motu-kairangi. 

Ka whakaaetia e nga niatua me te iwi ko Motu-kairangi. Te take 
i wliakciaetia ai taua ingoa, kaore rawa he wahi ahua raorao, mania 
he hnereerenga atu ki te haereere; tiro tonu ake i te po ko nga 
whetu, ko te marama, i te awatea ko te ra, ko nga kapua e tere ana i 
te rangi ; ko te moana i tahi taha, i tahi taha. Koia i kiia ai ko 
Motu-kairangi taua motu ; ko Whetu-kairangi te pa. 

Kei te taha mauru mai, kei reira tetahi roto hawai nei, he roto 
tuna whakaniokai, he mea man mai i roto i Te Awa-kairangi, ara i 
Heretaunga awa. Ko tenei ingoa ko Heretaunga no te taenga mai o 
Banginui me tona ope ki te toro mai i a Tara raua ko te taina ko 
Tautoki, ka noho, titiro ake he hiwi tahi taha, talii taha, he ngaliere- 
here. Katahi ka mea: — "Aue! Taukurira! Ki te kainga o nga 
tangata nei ; ko Heretaunga rawa pea to korua kainga kia whakamo- 
mon korua ki te nolio i konei." 

Ka mea atu a Tautoki : — " E ta ! Hei aha te kete tuwhera, kaore 
ki te kete ruru tau ana te niauri.*' 

Ana kupu a Ranginui mo te pai o Heretaunga, mo te marama, 
hei tawhiti ano te ope e haere mai ana ka kitea atu, kaore e ngaro. 
Ko Te Whanga nui a Tara he kino, he liiwi, he ngaherehere anake. 
Me ohorere te kitenga i te ope liaere mai o tawliiti. Na, ko ta 
Tautoki e ki ana ia lie pai rawa atu a Te Wlianga nui a Tara i Here- 
taunga, ora te tangata i te ope taua whakaeke, he ora i te kai moana, 
ka rahi te kai mahi a te ringa, e ora i te kai moana, i te manu o te 
ngahereliere, koia te kete ruru. 

Katahi a Hanginui ka ki atu: — " E tama ! Wailio te awa nei, a 
Te Awa-kairangi, hei whakamaliaratanga nio ta tatau korero ko 
Heretaunga." Koia a Heretaunga awa. 

Na, me lioki ano ta tatau korero ki te ope taua a Te Kopara o 
Mua-upoko ra. Ko te tino take i whakaae ai a Nga Rauru, a Ngati- 
Knanui ki taua amio i a ratau hei taua ngaki mate mo Pukehou, lie 
])irangi huia, kakahu Maori, mako mau taringa. Katahi ka haere 
mai ; kaore i pai tenei wahi o taku rongo; he aliakoa, i rongo au i 
haere mai taua ope ma runga waka, e wha nga waka, he waka taua 
anake. I u mai ki Porirua nei, ka noho i Papa-kowhai i te taha 
rawhiti o Ponrua nei, tatari atu ai ki a Mua-upoko. Ka tae mai a 
Mua-upoko i te wa e pua nei te kowhai. Ka tae mai ki te tauniata i 
Te Wharau, ka kite i nga ahi e kaka ana i Te Wai-hirere, i Te Aka- 
tarewa, i Uruhau, i Te Whetu-kairangi, i Pae-kawakawa, i Motu-haku, 
i Makure-rua, i Wai-komaru, te pa a Tukapua o Ngati-Mamoe, era 



Digitized by 



Google 



92 JOURNAL OF THE FOLYNESIAM SOCIETY. 

pa e irua; i te taha ki Te Rimurapa enei pa e rua. Ka ui atu a 
Tamatea-kopiri : — " Hei hea tatau o enei alii e kaka ake nei ? " 

Ka luea : — ** Waiho tatau i te ilionui i te wahi kaihora " — ara i te 
wahi e noho wehewehe ana nga tangata te noho. Ka wliakaae te 
taua nei. 

I te po o tetalii rangi iioa ake ka moemoea a Kauhika, he whaea 
no Te Rangi-kai-kore tenei, lie wahine nioemoe hoki tana kuia. Ka 
moe ia, e nolioia ana a Te Wharau e te tangata — ** I tahuna mai i 
reira te ahi nei tae tonu mai te niura ki ITruhau nei ; wehi tonu mai 
au, oho tonu ake nei au.'* 

Ka inea a Te Rangi-kai-kore: — **Me liaere he tangata ki Te 
Wharau nolio ai ki te taha ki te rawliiti o te matuaiwi, ki te 
tawhatitanga o te hiwi o te tau o Te Wharau nolio ai ki tahaki, me 
kore e tupono te marua a po a te kuia nei." 

Ka tonoa a Mohuia raua ko Kaipara, ka tae atu ki te wahi i 
tohutohutia atu ra e Te Rangi-kai-kore, ka noho i reira. Ka whanatu 
ka tairi te ra ki te huapae o te po, ka puta te ope nei e haere ake ana 
ki runga i taua hiwi i Te Wharau. Ka hoki mai aua tangata, ka 
mea, " He taua kei Te Wharau e matai ana ki te aliua o te kaka a 
nga ahi." 

Tonoa tonutia e Te Rangi: — "Haere ki Te Aka-tarewa, ki 
Uruhau, kia tukua nga wahine me nga tamariki ki Te Whetu- 
kairangi ; ka tono atu he tangata ki Para-ngarehu whakamohio atu 
ko te whakaariki tenei kei Te Wharau e matai ana mai." 

Ka haere a Mohuia ki Te Whetu-kairangi, ka haere a Kaipara ki 
Te Aka-tarewa tae atu ki Uruhau. Ko nga waka o te tangata 
whenua ka wliakawhitia ki te taha ki Motu-kairangi- Ka tiki n a ka 
tiakina te matuaiwi e tika ana mai i Te Wharau ma te taukahiwi e 
ahu whaka te tonga ana mai. Ka tonoa he tangata ki Pukeahu i 
runga ake o Hauwai ra, he ata marama hoki e ki ana te po. Ka kitea 
nga tangata e haere mai ana i te taha moana i -Kumutoto ra e haere 
mai ana. Ka hoki mai nga toro o Pukeahu, ka ki mai : — ** Kei te 
one i Waititi e haere mai ana te hiku, kei Kumu-toto nei a mua e 
haere mai ana. Me te uru ngahere te ruru o te tangata." 

Katahi ka nohoia tonutia i Kaipapa, i te wahi i te taha rawliiti o 
Hauwai ra tiaki atu ai, mehemea era e peka.ki tewhea pa ; ka kite e 
haere tika ana ki Uruhau te taua nei whakaeke ai. 

Ka rewa nga whetu o te ata, ka haere atu te tangata whenua o 
roto o Te Wai-hirere, ka tutaki atu ki nga tangata o Te Aka-tarewa. 
Katahi ka whakaputaia mai e te tangata whenua o roto o te pa i 
Uruhau ki waho. Kua taka tetahi wehenga o te taua ki te taha 
moana, ko tetahi wehenga i runga i te hiwi e awhi ana i te pa. Kua 
mohio tonu mai a Pakau ka mate taua taua i a ia, a me tona mohio 
tonu mai ki waho nga tangata o Te Wai-hirere, o Te Aka-tarewa 
whanga ana mai ki tona whakaputanga ki waho. Tera hoki a Tara 



Digitized by 



Google 



The Land of Tara. 93 

raua ko Tautoki kua eke kei ruuga o te liiwi i Orongo ; lie ingoa tera 
no Tamatea-ariki i tona taenga mai ki Te Whetu-kairaiigi. Ka piki 
ki runga i taua biwi matere ai ki roto i Te Whanga niii a Tara, ki 
tera niotu lioki. I Te Awa a Taia e kato aua a ** Takitumu," e aukaha 
ana, e pani ana ki te ware liouliou i nga kowliao, ka oti katahi ka 
haere a ** Takitumu" ki Arapawa, ara ki Te Wai-pounamu. Na Kupe 
tenei ingoa i tapa mo tera motu, nana lioki i kite te pounamu tuatalii 
ki Araliura i te taha mauru o tera motu. 

Na, ka eke ra a Tara, a Tautoki ki te hiwi i Orongo ki reira 
taupua ai i te paahatanga o te taua ki Uruhau. Ka marania ka kitea 
atu te taua i te taha takutai i raro iho o te pa o Uruhau, a kua puta 
rawa te taanga whenua o roto o Uruhau ki waho. I rangona atu ki 
te waha o Pakau e karanga ana: — "Huakina! Huakina!" Kua 
wehe etahi o te taanga wlieniia ki te ara i te takutai, kua ara 
hoki te pakanga a tera ara. Ka apiti atu hoki te taanga whenua 
o Te Wai-hirere. o Te Aka-tarewa ki te pa ki Uruhau. Ka 
karanga a Te Rangi-kai-kore : — ** Pakau E! Apitiria ! Apitiria!'* 
Ka rongo te taua, ka horo ki roto i te motu i te taha mauru o Uruliau. 
Katalii ka wliawhai ko te ara ki tatahi, ka mate a Te Toko, tetahi o 
nga rangatira o te ope taua i te whawhai ki Waitaha i te takutai i te 
kurae i te taha mauru o Te Awa-a-Taia. 

Ka po, katahi ka whakaaro nga tangata whenua, a Ngati-Hinewai, 
koi tahuri te taua ki te hauliake i nga purapura o a ratau maara 
kumara, kua pihi hoki te kuniara, nga tipu. Katalii ka hutihutia i te 
po, ka waiho hei ingoa mo Ngati-Hinewai ko Ngati-Hutihuti-po. Ka 
mutu ka whakawhitiwhiti ki roto i Te Wlietu-kairangi ; tae rawa atu 
a Te Rangi-kai-kore, a Pakau, a Te Piki-kotuku, nga rangatira o nga 
pa o te tua whenua, kua whiti nga wahine me nga tamariki, uga 
koroua ki Para-ngarehu e uoho ana mai. Ko te toa hapai rakau 
anake e noho ana i roto o Te Whetu-kairangi, ki tonu taua pa, e ono 
topu hoki a Ngai-Tara i taua wa nei. Ko te ope taua a Ngati-Ruanui, 
a Mua-upoko, e wha rau takitahi. 

I taua po ano ka tahuna a Te Toko, a Whakatau ki te ahi ki roto o 
Haewai, kei te taha mauru ma tonga o Te Rae-kaihau, taha moana. 

I te ata ka tahutahuna nga pa, a Uruhau, a Te Aka-tarewa, a Te 
Wai-hirere, me nga whare paenga maara katoa i Pae-kawakawa, me 
era atu maara katoa o te tua whenua. Ka tahuri te ope nei ki te niahi 
mokihi mo ratau hei whakawhitiwhiti mai i a ratau ki Motu-kairtingi. 
Ka rupeke mai ki Te Motu-kairangi, katahi ka whakaawhitia a Te 
Whetu-kairangi, te pa. Kotahi rau i Takapuna, kotahi rau i Kirikiri- 
tatangi, kotahi rau i Te Mirimiri, kotahi rau i te taha ki Kaiwaka, te 
roto i te taha mauru o Te Whetu-kairangi; i penei te whakaawhitanga 
i Te Whetu-kairangi. Ka mahia mai te rarauhe i te tua whenua hei 
whakapuini i te pa, kia tahuna ai ki te ahi in a puta te hau. Ka 
whakataetae ki te huri haere i nga powha rarauhe kia tata atu ki te pa ; 



Digitized by 



Google 



94 JOURNAL OF THE POLYMESiAM SQC/ETY, 

kore rawa i tata atu i nga taiigata o roto o te pa te werowero mai ki te 
manuka tarere mai i rnnga i nga puwhara o te pa. Tokowliitu nga 
tangata i mate i te tangata o roto o te pa, he mea tai*ere mai i intnga i 
nga puwliara. He penai te ahua o tenei rakan, he mea whakakoi 
tetahi pito, te pito i kaniwhatia, e tu ki te tangata ka whati atu ki 
roto i te tangata itia t\}, a ki te unuhia mai ka whati atu ki roto i te 
pito o te kaniwha, ka mau atu i roto i te tangata, a mate atu i tana 
kaniwha te tangata. 

E kiia ana ka noho te iwi nei i waho, ku puta te hau tonga me te 
ua i tetahi po, ka mate te iwi nei i te ua, i te maeke, ao noa te ra. Ka 
mate hoki i te kai kore, kua pan hoki nga kumara whakatipu o nga 
maara te kai e ratau, nga purapura o ro oneone. Kaore e taea atu nga 
kai o te nioana, te paua, te kuku, te pipi o Te Awa a Taia. 

Katahi a Tara ka ki atu ki ona toa taua : — " Kia awatea rawa 
apopo ka wliakaputa ki waho. Kia tokotovu ki te taki i te matua, nia 
muri e pehi atu ; kaua hei waiho ; e kore e roa e whawhai ana ka 
ngenge i te mate kai, i te mate i te niarangai hoki." 

Ka whakaae katoa te tangata o roto o Te Whetu-kairungi. I 
waenganui po e taka ana te kai, e kai ana ka awatea. Katahi ka 
whakaputaina atu a Te Whetu-kairangi ki waho. I te nui o te 
puaheiri o taua po tae noa ki taua ra i puta atu nei nga toa nei, kite rawa 
ake kua pau atu te ono rau takitahi ki waho o te pa. Ka horo te taua 
nei ki te taha mauru ki Te Awa a Taia, tae atu ki reira nga mea i tae, 
na te tai kato mai ka matemate ki roto i te wai etahi, ko etahi na te 
patu a te tangata whenua. Ka mate a Tamatea-kopiri, a Marohi, ka 
mutu nga rangatira ; ko tetahi o nga rangatira i mate ki roto i te wai, 
ka pae mai ki tahaki. Ki te korero he nui ano nga mea i puta, ara i 
whiti i te ia o Te A wa a Taia, te whakatere atu ki tetahi taha o te awa ; 
tae rawa ata te kai p«tu ki te taha o Te Awa a Taia kua whitiwhiti 
etahi, te nuinga. I mohiotia ki nga mea i mate, he kotahi rau tuma, e 
kiia ana no Mua-upoko te nuinga o enei i mate nei. Ka mutu tenei 
pakanga. 

Na, i muri o tenei pakanga, ahua roa ano te wa, ka tae mai a 
" Takitumu," ka noho i reira noho ai. I haere mai i Hokianga, i Muri- 
whenua, i a Ngapuhi ra. Kati, kua korerotia ake ra e au te nohoanga 
o Tamatea-ariki me te haerenga ki te Wai-pounamu, liaere a Tamatea, 
a Te Rongo-patahi, a Kohupara, a Puhi-whanake, a Kaewa, a Maahu, 
a, he maha nga tangata a Ngati-Waitaha- Ka hori atu a ** Takitunm,'* 
ka tae mai a Mapouriki, a Te Hoeroa, a Te Kahawai, na Whatonga i 
tono mai ki te toro mai i te whanau, mehemea kei te pewhea te noho. 

I konei ka wehe a Tautoki me tona iwi ki Wairarapa noho ai, ka 
tuturu ko Tara me ona iwi ki Te Whanga-nui-a-Tara noho ai, tae atu 
ano ki Wai-rarapa. Engari ko Tautoki ka tino wehe ia ki Wai- 
rarapa an ake, ahu atu ki Tamaki, ahu atu ki Te Keren ga o Mahuru, 
ka mutu mai ; ka tapahi ki roto ki Akitio awa, ka puta ki te moana 



Digitized by 



Google 



Th^ Land of Tara. 95. 

nui, ka haere mai i te takutai tae mai ki te WUanga-nui-a-Tara. ^a 
rere i roto o Heretaunga tae atu ki tona kauru, tae atu ki Te Bore a 
Malianga (i te pae maunga, taha ki te mauru o Pae-tumokai 
[Eeatlierston] e tata an^ ki Te Toko o Houmea), tae atu ki Kau- 
whanga, ka makere ki roto o Manawatu, ka aha ki uta, ka tae atu ki 
Kai-mokopuna, he mokopeke, ka tutaki tana rohe i konei. I a raua 
tahi ko te tuakana ko Tara me a raua uri, heke mai nei ki tenQi 
whakatipuranga e mau ana te mana o n^a uri ki nga wahi i mau i a 
ratau. Ko etahi wahi hoki he mea tuku ua raua, tae mai ki o raua 
uri me a raua mokopuna. 

Na, ko te taha hauaui*a, ka rere i roto o te awa o Heretaunga tae 
atu ki te kauru, ka mau atu ki Te Here-a-Mahaiiga, ka whati ki te 
mauru, Taumata-o-Korae, ka rere ki. roto i te kauru o Otaki, rere 
tonu i roto tae noa ki te moan a, ka whakawhiti ki Kapiti, rere mai ki 
te rae o Para-ngarehu, ka rere i te tahataha o te moan a, tae noa ki te 
ugutuawa o Heretaunga. Ka mau tenei wahi ki a Tara anake me ona 
uri me ona iwi ake. 

TE HAERENGA MAI O NGATI-MAMOE. 

I a Tara ano e ora ana ka tae mai a Ngati-Mamoe, ka noho ki Te 
Wlianga-nui-a-Tara. Ka tukua atu e Tara te taha ki Pahua, tae noa 
ki te moana, tae noa atu ki Te Rimurapa, tae noa atu ki Wai-pahihi e 
hangai atu ana te ngutuawa ki Arapawa, ka rere i roto o taua awa, ka 
piita mai ki te poho o Te Wharangi, he taukahiwi e takoto atu ki te 
takiwa ki Porirua ; ka rere mai ki te taha rawliiti nei o Te Wharangi^ 
ka makere ki roto i te awa o Waikqhu, ka aim whaka te rawhiti mai 
ki Te Whanga-nui-a-Tara nei, ka tae mai ki te kauru o taua awa, katahi 
ka ahu whaka te tonga, ka eke ki runga o te te hiwi ki Te Kopahou^ 
ka maro te rere i riinga i te hiwi ki te moana waitai taha tonga ki Te 
Hapua, taha mauru o te wahi e kiia ra ko Airana Pei (Island Bay), ko 
reira a Te Hapua ; ka rohe mai te wahi i tukua ki a Ngati-Mamoe i 
konei i taua wa. No te wa i heke a Ngati-Mamoe ki Arapaoa ka 
mahue te whenua, ka riro mai ano i a Ngai-Tara taua whenua katoa» 



TETAHI KOBBBO MO TE IKGOA NEI TE WHANOA-NUI-A-TARA. 

He mea tango mai i tetahi pukapuka i tuhia mai e Hone Wairere, 
Whanganui. 

Pari-whaiti, Oketopa 18-1911. 

E 1)oa, tena koe, ko te whatu ngaro i a koe e ngaro nei i te matau 
roa nei. 

E hoa, i rongo au he tangata whakamine korero koe ki roto i te 
rahu takai puni. Ana etahi mo roto i to rahu ; engari, ki te kore e 
marama i a koe aku relay e taea hoki te aha ; ka pa he whakairo rawa 



Digitized by 



Google 



96 



JOURNAL OF THE POLYKESiAM SOC/ETY. 



na aku dpuna, e taea te tauaki ki te ataata rau nui. Kati enei . . . 
. . . Ka haere mai a Tara me tana wahine, a Te TTmu-roimata, ki 
Heretaunga nei, tae mai ki te moana o Poneke nei, ka noho i te 
moutere i waenganui o nga wahapu e rua. Ka mahia tona pa, a Te 
Whetu-kairangi, kei te taha ki te ra; kei reira ano te puna wai 
whenua, a Te Puna a Tinirau. Ko Te Boto-kura te roto kei te taha 
mauru, he roto tuna na Wakanui 

No tetahi takiwa mai ka tapi te ingoa o taua moana e Te Umu- 
roimata, wahine a Tara. Ka mea atu ia ki tona tane : — *' £ koro ! 
Ko wai he ingoa mo te moana nei P " 

Ka mea mai a Tara : — ** Ko Tawhiti-nui." 

Ka mea atu a Te Umu :— " B ! Waiho tonu ko koe he ingoa." 
Koia Te Wlianga nui a Tara i aranga ai. 

Ka mea ano a Te ITmu-roimata : — '* Me waiho te awa e puta nei 
i te taha rawhiti nei ki te moana ko Te Au a Tane." 

Ka tapatapaia e taua ruruhi nga ingoa katoa o taua takiwa. Ko 
Porirua, mo te ruanga o nga moana te take, koia a Porirua. He maha 
atu nga ingoa a taua wahine i tapi ai. 

E hoa, he mihi tenei ka uruki atu na ki a koe— Tena hoe, te mata 
hapara o te kai taua. 

{To h$ continued.) 



Digitized by 



Google 



THE GREAT MURU. 



By a Taranaki Veteran. 



DURING the Summer of 1873 I was living at Opunake, being 
manager of the Opunake Company's flax mill, and being in 
charge of the Europeans in that part of the district, as they worked at 
the mill, matters. were brought more prominently to my notice than 
they otherwise would have been, and I had special facilities for seeing 
the full workings and effect of the muru of Te Namu kain^a, the 
greatest muru that had been known on that coast in the memoiy of 
the oldest Maori. 

Our mill and dwelling houses were situate on the south bank of the 
Otahi stream, and on the north bank, not far from the spot where Miss 
Dobie was afterwards murdered, was a Maori kainga^ or village, of 
some twenty -five or thirty houses, usually called Te Namu pa, though 
it was not the real Te Namu, famous for Wi Kingi's defence against 
the Waikato war-pai*ty in about the year 1835 ; that was an intrenched 
position near the mouth of Otahi stream, about half-mile more to 
seaward. 

One morning, about seven o'clock, while I was at breakfast a Maori 
woman rushed into the room crying out, ** Kahui has run away with 
Lydia." As our ideas of Maoii morality were not of a high order^ 
the news caused but a mild excitement, and I simply said, " Well, what 
then ? " The woman, astonished at my indifference, drew herself up, 
and in an angry tone repeated, "What then, what then, why there 
will be a big row, that's * What then,' " and stalked indignantly out of 
the room. I laughed, went on with my breakfast, and thought no 
more about the matter. About the middle of the morning, however, 
some of the leading men of the village came to Mr. Black — our store- 
keeper and paymaster — and myself, and told us, with grave and 
anxious faces, there was sure to be trouble over this escapade of 
Kabul's; that as the parties concerned were of high rank, the relatives, 
friends, and connections (other than those living in Te Namu) of the 
guilty parties, intended to muru Te Namu, that is, rob the inhabitants 
of the village of all they possessed as utu (payment) for Kabul's offence^ 
he being of high -rank in the village. One of the men who came to us 
with the news, was a regular old savage, a real old man-eater, and 



Digitized by 



Google 



«8 JOURNAL OF THE POLYNESIAH SOC/ETT, 

though he knew he would be robbed of all his belongings, he was in 
ecstasies at the prospect of the revival of a good old custom ; he danced 
around, he leaped and yelled, he quivered his hands and smacked his 
thighs, "Aha ! Aha ! " he cried, ** the Maoris of this day think they know 
all about muru, they think when a hoi-se or a cow, or a pig is taken, 
that is a muru. But you and they will see a sight that has not been 
seen on this coast for a hundred years ; you will see a real www," and, 
carried away by his feelings, he cried out, " O it is good to have lived 
to see this day," and quivered and shook and turned up the whites of 
his eyes and lolled out his tongue in a most diabolical manner. We were 
intensely amused at the antics of the old savage, but began to have a 
dim perception that a muru after the manner of the good old-times, 
might not be a matter for universal rejoicing. 

And now I will digress for a moment to explain what all this fuss 
was about. 

Kaliui te Kararehe was a young and handsome chief of high rank 
in the Opunake district, and lived at Te Namu. He was already 
taking a leading position in the district, being clever, energetic, and 
possessing a soft and winning voice, and persuasive tongue. Lydia 
was a handsome half-caste, tall and graceful, and fair in colour, and 
Kahui was madly in love with her, and she returned his passion. 
Kahui wanted to marry her, but having years previously been 
betrothed and married to a half-negress named Betty, a daughter of 
black Davis (who boasted he was the first tvkite man who came to 
Taranaki) the chiefs would not allow liim to have Lydia. Whilst 
Lydia — to use a Scotch expression — was married on to Aperama, a 
young chief of high rank living in Parihaka. Kahui's marriage with 
Betty was not a happy one, nor was Lydia's with Aperama, and the 
trouble culminated in Kahui's eloping with Lydia. As both Kahui 
and Aperama were of considerable consequence — prospectively — in tlie 
district, the chiefs within a radius of fifteen to twenty miles of Te 
Namu, decided the event should be marked in an emphatic manner, by 
reviving in its full power and consequences, the ancient law of muru. 
Hence the dread and down caste looks of the Maoris who came to us, 
and the exultant delight of the old savage. 

That afternoon those of us who were at times outside the mill, saw 
the Maoris from the nearest kainga (Matakaha — about a mile to the 
south of Te Namu) both men and women, troop past on their way to 
Te Namu, and after a time return, laughing gaily, and laden with old 
clothes, blankets, boxes, kits, eel-baskets,- cooking utensils — and in 
fact all the moveable possessions of the village on which they could 
lay their hands. Our Maoris — as I may term them — were left with 
nothing but the clothes they had on them, and their guns and some 
food they had hidden. After dark tlie men came to Mr. Black and 
Jianded him in their guus, asking him to hide them away, as should 



Digitized by 



Google 



ThB Grectt Kuru. 99 

any of the mnru parties find them they would at once be taken 
possession of, as pai*t of the spoil ; and tliey also requested him on no 
account to return them until the muru was ended, for said they, "We 
may be hard pressed, more than we can bear, and we might be tempted 
to use the guns, and that would be bad for us, and for you." 
Mr. Black accordingly promised to hide away the guns, and give them 
up to neither friend nor foe, until the trouble was over ; but he little 
knew the difficulties in store, nor the pressure he would have to resist. 

The next morning (the 2nd day) the Maoris from Umuroa, Nuku- 
te-apiapi and Waitaha villages, to the northward of Te Namu, came 
trooping in, and spent nearly the whole day in raiding every fowl, 
duck, goose, turkey, and pig in the vicinity of or on the lands 
belonging to the village ; not a feathered biped or grunter was left, 
and our Maoris began to look exceedingly gloomy. 

On the third day the Maoris from Taungatara, Punehu and Ouri 
villages — between Opunake and Oeo — came and took away eveiy horse, 
bullock, cow and calf they could find, not a hoof was lef t ; and we had 
to carefully muster our horses and working bullocks and pen them, for 
fear they also should be taken. Even the old savage looked glum, 
and tliought the affair liad gone far enough ; but the muru was by no 
means ended. By this time they had been robbed of every moveable 
thing they possessed, excepting the hidden food, and the guns in 
Mr. Black's charge. 

Early on the morning of the fourth day a runner came from 
Parihaka to Te Namu, with the news that an armed party had left 
Parihaka, and were coming to have their share of the muru, " But," 
said our Maoris, when telling us this, " what can they get, we have 
only our houses and our lives left, and they may be coming to kill us," 
and then they demanded from Mr. Black their guns back again, in 
order that if things came to the worst they could defend their lives and 
the lives of their wives and children ; it would not have been etiquette 
for them to leave their kainga until all was over. But Mr. Black 
wisely refused their request, as in the event of fighting we would be 
between two fires. 

We learned that the war- party — taua — would be at Te Namu by 
11 a.m. ; and as there was likely to be an exciting time, we closed the 
mill soon after ten o'clock, and went to Te Namu to see all we could. 

Not a Maori was there but those belonging to the hapu, it 
evidently was not correct for any but the principals to be present. The 
Maoris tried to persuade us to retire, as they knew not what would 
happen, and we might — accidently — be shot ; but we had come to see 
the show, and refused to budge. We — there were about 20 of us — 
posted ourselves on a slight eminence on one side of the village, where 
we could see all that would happen. Our Maoris occupied a slope 
just across a hollow which lay between us, and about twenty yards 



Digitized by 



Google 



100 JOURNAL OF THE POLYNESIAN SOCiETY. 

distant, the men in front squatting down, the women cowering behind. 
The side of the village towards Parihaka was left open. At about a 
quarter to eleven, we heard the Maoris coming along the track from 
Umuroa ; we heard an occasional sound of a chant rising or falling as 
they passed over hills or hollows, and at times there were volleys from 
their guns. 

They evidently desired to create an impression before their arrival. 
The track was hidden with a dense growth of flax, tutu and toetoe, and 
though we could liear we could not see. The noises continued until tlio 
war-pai*ty was within about one hundred yards of the village, and then 
there was silence ; a long, painful, anxious silence. 

I looked towards our Maoris, and the men were rigid and immovable 
as statues, but for an occasional quiver of a muscle, showing the 
intensity of the strain upon them, they might have been images of 
wood or stone. 

Presently, amidst the stillness, we could hear an occasional faint 
rustle, and judged the enemy was taking up its position near the edge 
of the scnib surrounding the village, and again there was intense 
silence. 

On a sudden the air was rent with screams, such screams as could 
emanate only from the throats of highly cultured female savages, and 
then two old hags, I verily believe the ugliest that could be produced 
in the district, sprang screaming, leaping and dancing into view. They 
were absolutely naked, and to add to their hideousness — if such a thing 
were possible — they had rolled themselves in the black mud of the 
raupo swamp. Each held in her liand a lighted torch, and each danced, 
screamed and reviled our Maoris in the choicest "Billingsgate" they 
possessed, and their vocabulary was an extensive one. 

They worked themselves up to a pich of frenzy, tore their cheeks 
and breasts with their sharp nails, until the blood ran down over the 
filth on their bodies ; and at last seeming to be able to control them- 
selves no longer, each rushed to a whare and shoved the blazing torch 
into the sides and roof ; the rau>po and toetoe, dry as tinder, blazed up in 
a flash, and they ran from luhare to whare ^ until every one in the 
village was in flames. A glance at our Maoris showed the awful 
strain upon them, they had drawn their blankets over their heads, and 
crouched forwards, their heads down almost to their knees. 

The two hags having done their worst retired, and not a sign of 
the enemy could be seen, nor a sound heard. But their silence and 
immobility was in a moment to change into active life; a ciy went up 
from one of our Maori women that an old bedridden Maori had been 
left in one of the whares, and was being burned to death. Men, both 
friends and iocs, rushed forward and reckless of burnt hands, arms or 
bodies, commenced frantically to pull down the whare, but it turned 
ont to be a false alarm, and back each went to their places, our Maoris 



Digitized by 



Google 



The Great Kuru. 101 

to their former position and attitude, and tlie enemy to their hiding 
places in the flax and scrub. The burning went on merrily, and in a 
comparatively short time nothing remained to show where the wkares 
had stood, but so many heaps of smouldering ashes, and then when the 
last frame had fallen, when the last tongue' of flame had died down, 
the war-party entered the village, headed by an old fighting chief 
named Tainihana, and took up their position — squatting, with their 
guns between their knees — opposite to our Maoris, a space of about 
ten yards separating them ; a beaten track running down the hollow 
lay between the two parties; the chief marched up and down this 
path in a slow and dignified manner. He was a stately but excitable 
old warrior, tattooed to perfection, he carried in his hand a splendid 
taiaha, which he used to give point and force to his speech ; he soon 
worked himself up to a pitch of frenzy, would march one way in a slow 
and dignified manner, turning at the end with a yell and spring, and 
ceme rushing back raving and dancing, waving his taiaha^ then on 
reaching the end of Ids self appointed beat, would turn, walk quietly 
back, tui*n witli a yell and spring at the end, and so on for nearly 
lialf- an-hour. At times he would violently revile the Maoris, at 
others would lament in pathetic terms the disgrace they — through 
Te Kahui — had brought upon, not themselves but, Parihaka. " Oh 
Farihaka, Parihaka," he once exclaimed, ''my heart grieves for 
you, and for the shame brought upon you by this people." At 
last, having exhausted both his subject and himself, he sat down at 
the head of and in front of his men ; and then (but it is hard for a 
mere Pakeha to credit this) the wives of our Maoris brought forward 
food they had cooked, and actually waited upon and feasted the 
wi'etches who had burnt them out of house and home. None but the 
enemy touched the food, our men sat perfectly still, and apparently 
unconscious of what was going on ; and when the food was consumed 
the enemy, without a word to our party or even a look at them, rose 
and departed by the way they had come. 

Everything had been done in perfect order, and in accordance with 
the best of their old traditions. 

When the last man had gone, and the sounds of their tramping 
had died away, our Maoris, with a sigh of relief, a long drawn 
inspiration, rose to their feet, and we went towards them; "Well," we 
said, " the muru is now over, the enemy has done his worst, and you 
have nothing more to fear." But one of the chiefs said bitterly, " No, 
it is not over, Titoko-waru, and his warriors fi*om Omuturangi, have 
still to .^ome." "But," we said "of what use will be his coming, you 
have nothing left ?" "Yes," was the gloomy reply, "we have our lives." 

That evening and the next morning were anxious times for all, the 
blood-thirsty, one-eyed Titoko was not the man to spare anyone until 
he had received full utu for his fancied wrongs. At about 9 o'clock the 



Digitized by 



Google 



102 JOURNAL OF THE POLYMESIAM SOCIETY. 

next moFuing (the fifth day of the Muru) the news came that Titoko- 
waru and a war-party had started from Omuturangi, and were on their 
way to claim their share of the mwru. Our Maoris were in despair, and 
practically demanded back their guns, but Mr. Black, knowing the 
danger of a collision with Titoko, that we as well as the Maoris might 
be swept out of the district or from off the face of the earth, was firm 
and refused to show where the guns were hidden. At last, at about 
eleven o'clock, when the excitement had reached fever heat, a 
messenger came galloping in, his liorse in a white lather, and cned out 
the danger was over. Hone Pihama of Oeo had intercepted Titoko- 
waru, and bribed him with a present of bullocks to turn back, that Tito 
had graciously accepted the bullocks in full payment of his claim, -and 
that the muru had ended, without the sliedding of blood. 

And whilst all this trouble was on, Kaliui and Lydia were spending 
their honeymoon in an old clearing some miles up the Oeo stream, and 
a few days after the muru was over they returned. Kahui all smiles 
and jubilant at having won his old flame, and Lydia dignified, but*a 
little shy of the new honours which liad come to her. 

Kahui was very pleased at having been able to confer lasting 
honour upon his hapu^ in making them the victims in such a splendid 
murUy the greatest on record ; the fact that they were homeless and 
lost all they possessed did not trouble him in the least, it was but a 
temporary inconvenience which would soon pass away, but the honour 
would last for all time. On meeting him I asked, '^ Well, Kaliui, what 
is to become of Lydia now? " " Lydia," he exclaimed," why Lydia is 
my wife, I bought her with the mwrM." " How can that be,*' I said, 
" Betty is your wife." " No, no," he replied, " Lydia is my wife now, 
hang Betty, slie can go and get another man." On making further 
enquiries I found that what Kaliui had told me was true ; tlie muru 
had dissolved the marriges of Kahui and Betty, of Aperama and Lydia, 
and had soleminized the marriage of Kahui and Lydia ; she became his 
legal wife according to Maori custom, and she remained iiis wife so 
long as he lived. 



Digitized by 



Google 



FURTHER NOTES ON THE HEO OF THE 
SOLOMON ISLANDS. 



By the Rev. C. E, Fox. 



MASITAWA AND HEO. 

WITH regard to the remark in my first paper to the effect that 
heo could have been derived from maaitawa, this was an error^ 
as, now tliat I have thoroughly examined the heo, I know they were 
not built up gradually in the way I suggested. The following 
description of the heo will make this clear. 

MSO. 
JSeo are usually quite small, about fifteen feet by ten feet, and 
very often two or three feet in height. It is only occasionally that 
one sees large heo. The largest I have seen is not more than sixty 
feet by foi*ty feet at the base, almost perpendicular, and fifteen feet 
in height. Along the coast they are made of large blocks of lime- 
stone, enclosing a rubble of smaller stones ; but in the bush they are 
built of reddish earth. I shall describe in more detail three heo : two 
on the coast at Tawaniora and Tawatana, and one in the interior at 
Mwanunu. 

1. THE HBO AT TAWANIORA. 

This is a solid stone structure built round a core of natural lime- 
stone rock under which there is a large cave. It is about ten feet in 
height, and about thirty feet from east to west, and slightly longer 
from north to south, so that it is an oblong lying north and south, 
which seems true of most at any rate of the heo. It is so built up 
round the soHd core as to leave a shaft running down to the mo^th 
of the cave, but the shaft had been filled up with large blocks by the 
last users of this heo, and I did not get down into the cave, which 
however, some of those with me had done in former times. On top 
of the heoy and to the south of the shaft, was a dolmen of two flat 
upright stones, and a large stone slab across, but this is now 
destroyed ; smaller stones placed on edge round this converted it into 
a kind of box, a receptacle for the bones of the dead. On the dolmen 
burnt sacnfices of dog and bonito were offered, but not of pigs or 



Digitized by 



Google 



104 JOURNAL OF THE POLYNESIAN SOCIETY. 

anything else. These were not private sacrifices, but for the whole 
clan. The dead man was lowered down the shaft into the cave, in 
which he lay on a bed. The body was carried frequently to running 
water; carried on the shoulders of a living man*; and the flesh 
stroked away from the bones. When the flesh was all gone the bones 
might be left in the cave, or (especially the skull or jawbone), might 
be placed in the dolmen. The latter is called hau sum, i.e., the raised 
or exalted stone. To the west of and alongside the ?ieo was an oha, a 
sacred building which women could not enter, and here first-fruits 
from the gardens were hung up, and boys lived for three years 
secluded, learning bonito fiehingc To the north was the men's 
dancing ground. The ?ieo had a guardian serpent. On the top were 
planted a sacred tree, tahi; and sacred dracaenas. At one corner was 
a sacred coconut, where private sacrifices of pigs were offered. On 
the hau suru was a sacred stone of nagi (flint) round in shape. 

2. THE HEO AT TAWATANA. 

The heo at Tawatana is also a solid stone structure but without a 
core of natural rock and a cave, and it is therefore more typical of 
heo in general. It is about fifty-five feet by thirty-three feet and runs 
north and south ; the height is not -more than six feet. It has no 
shaft running down into a cave, bat to the north of the hau suru is a 
shallow pit three or four feet deep, in which was a bed, on which the 
dead man was placed, and a roof was built over the pit — of thatch. 
Tlie hau suru is a fine one, the large flat slab being about four feet by 
two feet, with two large slabs at each end and smaller stones round. 
It contained chiefly skulls. The burial rites were the same as at 
Tawaniora. To the east of this heo is a large oblong cleared space 
called a hera, and this is normally found in this position ; but the 
word her a is also used for the heo itself, and seems to mean merely an 
oblong enclosed space. The oha was alongside, and I think on the 
west side of the heo, but I did not make a note of its position.f The 
first-fruits of the gardens were hung up in this oha as offerings to the 
snake spirit {htona) and sacrifices were offered on the hau suru to the 
Mona, Only people of the Araha (chiefs) clan were buried on these 
two heo. The Araha are connected in many ways with the sun, and 
when one of them dies his widow often ** marries the sun.'* 

3. THE BBO AT MWANXmU. 

This is a typical bush heo, but larger than most. Its sides are 
almost perpendicular, and, since it is now much decayed, it cannof 
formerly have been less than fifteen feet high, judging from the 

*' Ak in Torres Ifilands. 
t„Po88ibly it was in the hera. 



Digitized by 



Google 



Further Moie$ on the Heo of the Solomon Mande. 105 

aoooimts of those wlio remember it when in use; but it is now 
perhaps about twelve feet high (I could not measure the heo exactly). 
It is almost square, about fifty feet by fifty feet. On the top there is 
the usual pit, now fallen in, but said to have been about eight feet 
deep, and widened at the bottom to form a cave, in which the dead 
were buried. I saw no hau suruy and could not find out if one had 
formerly existed. Only araha men or women were buried in this A^o, 
and by the mouth of tlie shaft was a small liouse, for the husband or 
wife of the dead person, who lived there for ten days till the mourning 
ceremonies were concluded. The shaft was lined to-prevent the earth 
falling ill. The dead person was carried on the shoulders to running 
water, and bathed (on the sliouldevs of the living) by a man who 
stroked o£E the flesh with the palms of his hands. Thei*e was no oha, 
and sacrifices of first-fruits were hung on the branches of sacred trees 
growing on the heo. The heo stood right in the middle of the village 
with houses all round it ; not some little distance away as in most 
coast heo. 



Digitized by 



Google 



VERY LIKE SCALP-TAKING. 



By S. Percy Smith. 



THE following account of a Maori custom in war was told by an 
old Maori friend who was well up in their old customs. It is 
worth preserving as it has not been mentioned elsewhere so far as I 
know. 

In a battle where the enemy is routed and flees, the toas or braves 
of the victorious party dash forward in pursuit, each bent on killing 
as many of the enemy as possible. Many a warrior would, on over- 
taking his victim, give him a blow on the head with his mere, or 
other weapon, and leave him there for others to finish the work, whilst 
he dashed onward to secure others in the same manner. After the 
fight, and when the victorious party were in their camp, it was 
customar}' for eacli of the warriors to boast of his deeds (called 
korero-tvkakatu). Each would say, **I killed so and so with my 
weapon, etc." and describe all the blows given and received, with 
minute details. After the victoiy of Tai-karamu, fought somewhere 
on the East Coast, all the victors were boasting in tlie above manner, 
some declaring they had killed more of the enemy than anyone else. 
A warrior named Tu-kokoru then stood forward, and untying his 
girdle let it fall to the ground, when it was seen that there were a 
great number of koukou, or top-knots sucli as each warrior wore, some 
with feathers stuck in them, others witliout. He said, "You are all 
boasting of the number you have killed. Behold my hokowhiti,*' i.e., 
his seventy koukou. Then the top-knots were counted, and there were 
found to be seventy. So all agreed that Tu-kokoru had killed the 
most men. On striking down his man each warrior would lay hold 
of the koukouy and using a chip of obsidian, wliich he carried in his 
belt, cut off the top-knot below the lasliing, stick it into his belt, and 
then dash on after the fleeing enemy, all the time repeating his 
charms to cause the enemy to stumble and fall. In Tu-kokoru's case 
these charms were named ** Tu-mania'* and **Tu-paheke.'' Another 
name for a similar charm is " Pa-whakaolio." 

Tu-kokoru lived five generations after tlie well-known ancestor 
Pou-heni, and therefore seventeen generations ago, or in the end of 
the fifteenth century. 



Digitized by 



Google 



RANGl-HUA-MOA. 

A LEGEND OF THE MOA IN WAITEMATA DISTRICT, AUCKLAND. 



By Geo. Graham. 



IN the "Transactions of the New Zealand Institute" for 1916, 
page 425, appears an article by Mr. T. W. Downes, dealing with 
the vexed question of the probable period at which the Moa became 
extinct in New Zealand. After reading it I at once turned to an old 
memorandum book, wherein I had made an entry in 1910, of a legend 
in respect of the Moa in these parts — my informant being old 
Mereri. The old lady and myself had been talking of the ancient 
times, as was oft our wont, she being my informant of much ancient 
lore in which she was well versed. It was from her I obtained a 
version of '* The Korotangi Mytli," vide this *' Journal," Vol. XXVII., 
page 86, where also appears her whakapapa on page 89, and which is 
supplemented by a further ivhakapapa at end hereof. As is often the 
case in obtaining this kind of information, it was a casual question of 
mine as to the meaning of her daughter's name, Eangi-hua-moa, tliat 
led her. to give me the following: — 

RANGl-HUA-MOA. 

** This daughter of mine I named so at the request of my relative 
of Te Aki-tai hapu. It was Taka-anini who asked me to so name her; 
it was the name of an ancestress of us both. Such was the Maori 
custom ; that naming gave the right to the guardianship as to 
marriage in due course, of children so named. Hence was my 
daughter Eangi-hua-moa named, and she ultimately married a younger 
relative of Taka-anini, when she came of age. Thus were inter-tribal 
relations cemented, and marriages into foreign tribes discouraged, 
thereby securing lands and other heirlooms, and preventing a 
depletion of tribal-membership. 

Rangi-hua-moa was named after the mother of Te Ika-maupoho, 
who married Te Tahuri ; these were the parents of Kiwi Tamaki, 
from whom Taka-anini is descended (as shown in whahapapa). This 
is the meaning {putake) of that old ancestral woman Eangi's name. 
She was at Te Pani-o-poa-taniwha at the head of the Parenioremo 



Digitized by 



Google 



108 JOURNAL OF THE POLYNESIAN SOCIETY. 

creek, a branch of the Wai-te-iiiata. Her parents were there catching 
and drying eels for a feast. Such vas that place at Paremoremo^a 
place wliere at certain seasons they resorted to catch eels, birds ; also 
to collect the edible flower of the kiekie, then plentiful in those parts ; 
and the leaves thereof for mat-making. 

On the day Kangi was born a nest {kohangd) of Moa eggs was 
found by Huri-aka (a mokai or slave of Hine-korako's). The eggs 
provided a feast; the last time such a feast was held, for no Moa 
eggs were found after tliat time. In a few days the child was called 
Rangi-hua-moa (the day of the Moa eggs). Tlie tracks {ara) of the 
bird can still be seen on the ridges it formerly frequented, for it 
avoided the gullies and deep forests, and the ridges lightly forested 
and open places were its habitat {nohoanga). There on the ridges 
{hiwi) of the liill top of Te Pani-o-poa-taniwlia, is the old poka (pit) 
wherein that bird was trapped. A Moa having been located on the 
ridges, a party of hunters gradually drove it along towards the pit ; a 
party also approached from the opposite direction until they drove that 
bird into the pit where it was easily killed with spears and clubs ; for 
it never attempted to escape into tlie gullies, nor, if slowly followed, 
would it run back through the line of hunters. The reason it was so 
killed was the great speed at which it would run, if too quickly 
chased, as also the fear of its power of kicking {ichana). 

When our people first saw the foreign birds brought by Governor 
Grey to Te Kawau,* we called it the Moa, and my cousin, Te Hemara, 
made a speech to those birds and cried, and we all there cried, for we 
remembered those old proverbs and laments concerning the past, 
which likened the disappearance of our dead parents and ancestors to 
the extinction of that bird, the Moa. 

In some parts they hunted the Moa with the long spear {tao-roa)^ 
similar to that with which the pigeon was speared. In districts where 
there were swamps, the bird was driven therein. In Kaipara is a 
swamp known as the Te Toreminga-Moa (near Te Kapoai at 
Helensville). A name our people had for the Moa was Te Manu- 
pouturu (the bird on stilts), because of its peculiar walk, hesitating 
and awkward, like a person walking on stilts, and looking round 
every now and again before walking further. Te-rau-a-moa, at 
Pirongia, is not a name-place given after the Moa itself. At that 
place a battle was fought, and the slain lay about like the bodies of 
so many Moas after a Moa Imnt, hence that name, * The hundreds 
like unto the Moa.' " 



* Sir George Grey had a flock of Emu imported aud placed on the Eawau 
Ltlandy where they are still found, and gave several to John Reid, of Motutapu, 
where is also a flock of Emus still to be found. 



Digitized by 



Google 



Rangi-hua-moa. 109 

Such was Mereri's account of the Moa. I regret I did not keep a 
record of several proverbs and waiata slie recited, hut my recollection 
is that they were on very similar lines to those otherwise on record. 
I recollect, however, she told me that garments of the feathers of 
tlie Moa were prized because of their warmth, and that the feathers 
of the Kiwi and Moa were woven into the flaxen garments in a 
different manner to other gaiments, all the feathers being so woven 
that they stood in tufts outward, and not so that they hung downwards, 
as with the pigeon and other feather garments. 

The pit on the ridge referred to, I liave often examined with 
curiosity, it is about twenty feet by eighteen and perpendicular at 
the sides, and about five feet deep. There is nothing like it in the 
neighbourhood. Several years ago crossing from Okura to Pare- 
moremo, I came across a similar pit on the ridges near the Pukeatua 
trig station. 

The explanation of tlie naming of the ancestress Rangi-hua-moa, 
after the last recorded find of Moa eggs in these parts, is interesting, 
and I see no reason to doubt the truth of the legend. All Maori 
names of persons and places are derived from some such domestic or 
tribal event. I am inclined to place the time of Bangi's birt)i about 
1660, allowing thirty years* for each generation to the present time. 
Apparently the feast of Moa eggs was a " red letter day," from which 
we may conclude that at that time the Moa was at least in the district 
of Waitemata-Kaipara districts, on the verge of extinction ; otherwise 
the event would hardly be remembered as the birthday of an 
important chieftainess. 

I have never been able to discover any remains of the Moa myself 
in these parts, except those relics known as Whatu-moa (Moa stones) 
and Tutae-moa (Moa droppings), which are the crop-stones of the 
bird and are found in most districts all over New Zealand. Tliese 
stones are particularly plentiful on tlie ridges along whicli, as Mereri 
states, are still to be found what are certainly tracks, and where it 
was not conceivable that there ever was any very extensive human 
traf&c. These tracks are still fairly well defined among the scrub, 
and are known here, as elsewhere, as Ara^moa, and resemble 
abandoned sheep and cattle tracks. Old natives assure me that they 
existed in their youth ; before ever cattle came into these districts, 
and it is along these tracks that the so-called ** Moa-stones " are very 
numerous in small patches of a dozen or more. 



* Since 1893 the Society has adopted twenty-five years to a generation (not 
thirty years) in calculating dates. To make Mr. Graham's date conform to this 
rule, it should read 1676.— Editob. 



Digitized by 



Google 



110 



1660 



JOURNAL OF THE POLYNESIAN SOCIETY. 
HINE-KORAKO. 



Rangi-hnamoa ^ Hua 

(of Te Aki-tai) 

I 



1690 Te Tahuri = Te Ikamaupoho 
1723 Kiwi Tamaki 



1750 



Rang^matom = Moenehu 



= M(> 



Te Ahiwera 

I 



1790 PukiteHau=TeTihi 



1820 
1850 



Ihaka Taka-anini * 

Wirihana and others 
living at Mangere now 



Uronga-tapu 

Te Anini Oha-ki 

I I 

Te Hemarat Mereri 

(Descendents Rangi-hoamoa 
at Mahurangi) | 

Waipaia 

Te Ata and 
others (18 years old). 



The above Whaka-papa agrees very closely with such parts thereof as appear 
in Fenton*8 ** Judgments," and is supplementary to Mereri's Whaka-papa given 
in Volume XXVII., page 89 of this ** Journal "—see ** A Legend of Old 
Mahuraug^.'* 

* A fine, handsome old chief in 1860, who lived near Papakura, Auckland. He 
was supposed to convey information to the enemy Waikato during the Maori war 
of 1863, and was consequently interned at Mangere. — ^Editob. 

t Another very fine chief of the Ngati-Rongo tribe living at Puhoi and 
Mahurangi in the early sixties of last century. — ^Editob. 



Digitized by 



Google 



THE ACCOUNT OF KUPE AND TAINUI. 



By Geo. Graham. 



[It is obvious that this Hccount of Kupe refers to the second of that uame, and not 
to the original discoverer of New Zealand, who flourished some ten generations 
before Toi, mentioned herein. See our ** Memoirs,'' Vol. IV., for the account 
of the original Kupe. — £ditob.] 

THE following is a nhort account of some incidents connected with 
the arrival of Kupe in New Zealand, his return to Hawaiki^ 
and the subsequent arrival of the canoe **Tainui" at Tamaki, 
Auckland, and of the early connection of that district with the 
immigrants who came in that canoe. 

KUPE's visit to new ZEALAND. 

It was Kupe who first came liere to New Zealand bent on 
exploring lands he had beard of from his elders, and known to still 
more ancient times. Such is the story as told me by my elder Tati 
Wliarekawa. 

Kupe and his people discovered people at various places. These 
people were the Mamoe, the Turehu, the Tahurangi, the Poke- 
pokewai, the Patupaiarehe, the Turepe and the Hamoamoa. They 
hved on tlie fronds and berries of the trees, and the roots of the earth. 
They were expert in preparing such foods, and in snaring and 
spearing the birds in forest and fish in stream. They also prepared 
food from the tender parts of the nikaUy the tikoukou, the para and the 
mamakn (tree ferns). Another name that people were called by was 
Te Tini-o-Toi-kai-rakau (the multitude of Toi, eater of trees). Toi 
being an ancestor of a section of that people. They dug the roots 
with long ko (spades), an implement unknown to the Maori before we 
came to tliose islands, and found those people just as Kupe had 
described tliem. Kupe was attacked by, and in return attacked those 
people of Karioi, near Raglan, and Aotea on the West Coast. These 
people were the Ngati-Matakore so-called, not the tribe of that name 
now living here in this island, who descend from us of '* Tainui." 

Now Powhe-te-ngu,* his slave, he left with some companions at 
Aotea. His daugliter he had left in that other island (South Island) 
at the place he called Taoiiui-a-Kupe (tlie eastern head of Queen 
Charlotte Sound), where he had first landed in that island, and at 

♦ Is this not Po-whete-ugu ? Editor. 



Digitized by 



Google 



112 JOURNAL OF THE POLYNESIAN SOCIETY. 

Raiigitoto, D'Urville Island (Cooks Straits). He left Po to look 
after this island and his daughter to look after that other island ; for 
he knew others would come and claim tliose parts if he did not thus 
name places and leave people and signs of possession before he 
returned homeward. Kupe ordered Po not to follow, but to remain 
in possession, and to ensure this he left the surging seas off the West 
Coast at Marowhara. 

POWHE-TB-NGU ATTEMPTS TO RETURN TO HAWAIKI. 

Now Po lamented after the departure of Kupe, nor did he remain 
willingly, for he feared the hostility of the original people of the land. 
Therefore he built a canoe to return and follow Kupe to Hawaiki. 
He called that canoe the "Rewa-atu." He thereby disobeyed the 
injunction of his head chief Kupe. He gave out a report that the 
canoe was for fishing, but it was not so, it was for the purpose of 
returning to Hawaiki. When the canoe was complete, he and his 
companions secretly loaded it with stores of fern-root and started 
forth. But the kawa (ceremonial incantations) performed by Kupe to 
prevent Po returning caused the seas to surge and the winds to blow 
adversely ; Po's canoe capsized and he and his companions perished. 

That canoe was turned into stone, also that man, and they may 
still be seen at theWahapu-o-Aotea(harT)Our entrance of Aotea, West 
Coast, North Island.) Hence the i3roverb applied to disoT)edient 
servants who do otherwise than as commanded by their masters, ** Ko 
te mahi o te uri o Poivhe-te-NgtC^ (" The doings of the descendants of 
Powhe-te-Ngu.") 

THE BUILDING OF 'iTAINUI." 

The multitude in Hawaiki heard of this Island on Knpe's return ; 
then it was that Whakaoti-rangi asked for the mauri (emblem of 
divine assistance) of Puanga, which was the rori of the house of 
Uenuku, and of his father Memeha-o-te-rangi. Her request was 
granted, and the tohunga Rata-o-Waliie-roa undertook to build a 
canoe to come hither from Hawaiki. The three sacred adzes of 
Hine-tua-hoanga were brought, and their names were : — 
" Hauhau-te-po " — -the felling axe 
** Paopao-te-ra " — the splitting axe 

" Manu-tawhio-rangi " — the smoothing or finishing off axe. 
Thereupon Eata, after proper ceremonial, began the building of 
'' Tainui." 

(Then follows the account of the felling of the tree, and the 
building of the canoe " Tainui " therefrom ; much on the lines of the 
usual accepted narrative.) 

When the canoe was finished, and was being tried in the water, it 
was Marama-kiko-hua who called out to Hotu, ** E Hotu, your canoe 
is " Tainui " — hence she was so called. 



Digitized by 



Google 



The AeGount of Kupe and TainuL 113 

Ma4iy sacred treasures were brought in "Taiuui," including the 
abov^e adzes. 

"tainui" voyages from hawaiki to rarotonga. 

From Hawaiki they came witli ** Te Arawa " canoe.; Ngatoro-i- 
rangi was the " Tainui " navigator. 

When they arrived at Rarotonga [sic] the people of that place 
were urged to come cdong also, that is to say, the people called Te 
Aitanga-o- Whakaahu, younger brotlier of Puanga ; but those people 
said "No" — they would not agree to leave their ancestral home and 
come liither — so they were left behind ; also Eakataura of the 
** Tainui" crew, because of his thievish liabits. Eiu-ki-uta W6is now 
the navigator, because Tama-te-kapua liad taken away on his canoe 
the "Taiuui's" navigator Nga-toro-i-rangi, also tliat man*s wife 
Kea-taketake. 

"tainui" arrives at whangaparaoa. 

When the ** Tainui" canoe approached the shore at Whanga- 
paraoa, it was surrounded by the sea monsters who were led by their 
leader Makawe-nui-o-rangi . To appease these monsters, they recited 
the appropriate incantations known as the tu& and the takamate. To 
prevent a disaster on lauding in the surf, the tohunga Riu-ki-uta first 
jumped into the water, recited an incantation to appease the monsters, 
plucking forth a lock of his hair, he threw the same together with his 
red feather plume {rau-kura) into tlie sea. 

The sea monsters then departed, and they landed in peace. Tliey 
performed many other appropriate ceremonies before they wandered 
over the land. For the deities of the land and forest and the rivers 
thereof would not be forgotten, and the land must first be freed from 
tapu {whaka-nqa to be made fit for common use.) 

RAKATAURA's SIHATAGEM. 

To their surprise tlien stood forth that thief Bakataura ; he had 
got there by a stratagem. That man had turned himself into a rat, 
and had hidden in the interior of the canoe. Hence the proverb 
applied to that man, ** Tlie new rat twice killed," an expression now 
applied in contempt when a person is quarrelling with his descendants. 

HAPOPO'S FEATHER PLUMES. 

As to the story of Hapopo, of the "Tainui" crew, it was he who 
first threw his red feather plume into the sea and first plucked the 
bloom of the pohutukawa (Christmas tree) to replace the same. This 
plume of feathers drifted ashore, and was found by Mahina ; hence 
the name of any treasure cast up by the sea, " He kura-pae na 
Mahina,^'' (The cast up treasure from the sea, of Mahina.) Our 



Digitized by 



Google 



114 JOURNAL OF THE POLYNESIAN SOCIETY. 

forefathers called all red stone on the sea-coast " Nga Tai-apa-kura,^^ 
another name being " The great plume of Hapopo." The hira of 
Hapopo was treasured and kept by the descendants of the "Tainui's" 
crew for many generations, for it revived the memory of the ancestral 
home at Hawaiki. All the other plumes thrown into the sea, in order 
to replace them by pohtitnhawa blooms, were lost, only Hapopo's was 
found and preserved by the descendants of Mahina, at Kawhia. It 
was the custom to bring it forth in order that it might be lamented 
over when friends long absent arrived from diistant parts, or when the 
people were assembled to weep over the dead. Such were the 
treasured things {taonga) of old and the customs of our olden days. 

" TAINin's " VOYAGE NORTH TO HOKIANQA. 

** Tainui *^ arrived at Whanga-paraoa on the East Cost of the Bay 
of Plenty ; at that place occurred the incident of the whale claimed by 
^*Te Arawa" canoe party whom they met there. "Tainui" then 
explored northward till she rounded Muri-whenua (Land*s End, i.e.. 
North Cape) and entered Ilokianga Harbour. From there they 
returned, not being able to proceed further southward because of the 
great seas left there by Kupe as before-mentioned. 

** tainui" explores hauraki. 
Tliey then returned to Hauraki and came to the inoutli of tlie 
Piako River. Here they sounded the water to try the depth, the oar 
stuck fast. The canoe drifted on, leaving the oar standing up in the 
mud. Hence the name of that district even to these times — ** Te 
Hoe-o-Tainui " — Tainui's oar.* This oar was left as a token to those 
coming after them of Tainui *s previous claim by discovery, and an 
act of taking possession. Now this oar was preserved by the 
descendants of Mani-tu-ahu until recent times. From Piako the 
" Tainui " came to AVharekawa on the west side Hauraki Gulf. Here 
Marama left the canoe with some other companions and some slaves, 
and journeyed onwards by land to that river at Otaliuliu, so called 
now-a-days, but in olden times known as Whangai-makau. 

** tainui " EXPLORES WAI-TE-MATA. 

** Tainui " voyaged on from Wharekawa to Wai-te-mata, or 
Auckland Harbour, They rested at Takapunaf (North Head) on the 

* The name of a place a long way up the Piako River, not far from Morins- 
ville. — Editor. 

t Takapuna. — This I understand is not the actual name of North Head, 
Auckland Harbour, which is no doubt forgotten ; but the name of a still existing 
spring of water on the western slope of the hill, near Mr. Watson's home there. 
There are several places so named in New Zealand. 



Digitized by 



Google 



The Account of Kupe and Tainuii 115 

beach tliere below the Kurae-o-Tura * [sic] . Here seeing the flock of 
sea birds coming from the west, they thought thus, ** Yonder is the 
sea — ^let us go and explore." So they departed hence to the westward. 
Te Horoiwi remained with the people of that land. Hence tlie name 
of the lieadland pa eastward of Wai-pareira (St. Hellier's Bay) called 
Te Pani-o-Horo-iwi (Horo-iwi's liead), and the name of that kapu 
(clan) of his who dwelt thereabouts. 

TAI-KEHU SENT TO EXPLORE MANUKA. 

And so they sailed on to tlie Tamaki river and rested there. 
Tai-kehu and some others went on overland to explore. They 
proceeded to various parts of the western district. They found the 
waters of that sea called the Manuka (now Manukau) teaming with 
fish, and caught many mullet (kanae). So great was the plenty of 
mullet, they could catch one with eacli hand at a time. Hence the 
proverb applied to the ** Tainui " descendants of Tamaki as a motto 
(pepeha) : ** Nga potiki toa a Taikehu,'*'* (The brave young children of 
Tai-kehu-nui). He was also the ancestor from Hawaiki of those 
tribes of Takapuna and surrounding districts known as Ngati-Tai. 

"tainui" voyages from manuka to kawhia. 

'I'ai-kehu and his party returned to Tamaki and reported on their 
discoveries of tlie seas of Manuka, and of its opening into the ocean 
to the west. So they decided to drag that canoe ** Tainui " into that 
sea over the Otahuhu isthmus. Now Marama had arrived from her 
overland journey, hence they called that place on the Tamaki, near 
Otahuhu, " Te Wliangai-makau." (The awaiting for one loved.) 

From here ** Tainui" was dragged into Manuka after the proper 
ceremonies to counteract the effects of Marama's loose conduct during 
her absence from her party. 

Then ** Tainui ** floated on Manuka and passed on to Kawhia. 

But that story is well-known ; I will not here recite it. 

'^tainui" pioneer shelters at tamaki and hauraki. 
Now the people of ** Tainui " canoe who remained or came to live 
at Waitemata and Tamaki were : Riu-ki-uta, Pou-tukeka and his 
wife, Hapopo and his wife, Te Uhenga and his wife, and Hautai and 
his wife. These lived with the people of those parts, and their 
descendants were all important chiefs of the Tamaki tribes, who were 
thus connected with relatives at Kawhia and Hauraki. Hence it was 
that Titahi of Taranaki, and also in later times Maki, came hither 
from the south to see his ancestrally connected tribes, but he 

* Kurae-o-Tura. — Tura's headland — a pn in olden times ; it is now ahnost 
excavated out of existence for roadway and quarrying purposes. It is near the 
Ferry Coy.'s workshops (Devonport). Tura's identity is not now possible to 
ascertain. 



Digitized by 



Google 



116 JOURNAL OF THE POLYNEmN SOCIETY. 

quarrelled witli and fouglit and conquered them. Hence it was also 
that Hotunui, insulted because of a false accusation of theft, came 
from Kawhia and lived witli his relatives of Tainui at Wharekawa; 
his son Marutuahu, who afterwards came to seek him, also settled 
there. Hence the tribal name Marutuahu applied to all the kindred 
Hauraki people. Tliat is the name whicli links those people with 
Kawhia of ** Tainui." 

ORIGIN OF THE NGA-OHO TEIBE. 

Now the old tribes of Taniaki were known as Nga-oho. This is 
tlieir whakapapa : — 

Hamoauioa was an ancient ancestor of those people ; he was born 
of Kangawheuua, his father being a god. Hamoamoa gave the name 
to those ancient people who were the tangatci'tohenua of those parts — 
Tamaki and Waitemata. 

From Hamoamoa was descended Maheu, 

Maheu = Tahiiiga (of " Tainui ") 

Bakei-ora 



Raurangi Oho-mata-kamokamo 

These two brotliers quarrelled over their tribal areas at Barotonga 
(? Mt. Smart, near Auckland) ; and a long war resulted, each having 
many thousands of followers. Hence they are known as ** Nga tamariki 
hiktiroa a Rakei-ora.^'* (The long-tailed cliildren of Rakei-ora.) A 
hikn-roa being a name given to a chief who had a large following. 

So great was the loss of human life, and the destruction of villages 
and cultivations, that this warfare has given rise to the well-known 
proverbs, ** He mara-pungarehu " (a cultivation become an ash-field, 
i.e., a fruitful area become a desert) ; also ** Te Pokuru nut a Ruarangiy 
me nga namunamu o Hurihuri" (the great stopped-up flow of 
Ruarangi — even as the sand — flies of Hurihuri." (The loss of lives 
being likened to the flood of waters and the numbers involved as 
numerous as sandflies on the sen-beach at Hurihuri.) 

After a long and devastating struggle, Oho drove Ruarangi and a 
remnant of his followers into the Owairaka pa (Mt. Albert). Here he 
besieged them. Ruarangi and his people eventually evacuated the pa, 
escaping by a secret cavern, and went away to Waitakerei. 

Oho remained the victor, and gave his name to that olden people 
of Tamaki known as Nga-Oho. (The actual distinction from the 
Tainui people and Ngati-Whatua people of the north, as well as 
Wai-o-hua, is hard to define, owing to the extensive intermingling 
by marriage and so on — they were indeed a much mixed people.) 



Digitized by 



Google 



THE VISIT OF DENTRECASTEAUX 

TO THE NQRTH CAPE, NEW ZEAIiAND, 

IN MARCH, 1793. 



IT will prove of interest to members of our Society to learn 
something of t)ie visit of tlie French Expedition in search of 
M. de la Perouse, who had sailed from France in 1787, on a voyage 
of discovery, and had disappeared completely after liis departure from 
Port Jackson for the Solomon Islands and those parts. Many years 
afterwards it was discovered by the Chevalier Dillon that Perouse's 
two frigates, the ** Boussonlle " and the ** Astrolabe,'* had both been 
wrecked on Vanikoro Island of the Santa Cruz or Queen Charlotte 
Group, lying to tlie south-east of the Solomons. 

On the 9th February, I7ai, the **Assembl6e Constiuante " of 
France addressed the King asking liim to dispatch two vessels in 
search of Perouse, and in consequence the frigates ** Recherche," 
under the command of General Dentrecasteaux, and the **E8perance " 
commanded by Major Hi ion Kermadec, sailed from Brest on tlie 28th 
September, 1791. 

Tlie history of this voyage has been written by the celebrated 
naturalist, ** Citoyen" Labillardi^re, and was published in Paris, in 
two quarto volumes and a volume of plates, in the year Vlll. of the 
Republic of France (1780). 

Tlie doings of the expedition during the earlier part of the voyage 
do not concern us just now, so we commence our translation at 
Chapter XII. of Vol. II. The two vessels were at that time in 
Tasmania, and had been there some time exploring the coasts, along 
which they left many names which are in evidence to-day. Of course 
this wajB before Tasmania was colonized. 



** We made sail from the Bay of Adventure at 9 a.m. on the 10th 
Yentose (March), 1793, and were forced along by strong breezes from 
the south-west, and were not long in passing Cape Pillar, })ehind 
which we saw many fires lighted by the Savages. We then directed 
our course to the north, we were about eight "Myriametres" in the 
offing, leaving to the west the Bay of Oysters, afterwards we laid our 
course foi* the Friendly Islands. 

The 22nd, from daybreak we sighted the isles named Les Trois 
Rois (The Three Kings). At 8 a.m. we were in longitude 169® 56' 
west (of Paris), when we sighted to the north at a '* demi-myriametre " 



Digitized by 



Google 



118 JOURNAL OF THE POLYNESIAN SOC/ETK 

of distance, the middle isle of that group, and we determined the 
latitude to be 34° 20' south. We noticed three principal rocks (isles) 
of a moderate elevation, placed nearly on the same parallel of latitude 
and not very far apai-t, and surrounded by many other rocks veiy 
much smaller. In spite of the haze that arose we saw others towards 
the north that form part of the same group. They appeared very 
aiid, and we presume they are not inhabited. Nevertheless, a strong 
column of smoke above the most eastern isle announcing the presence 
of the Savages. No doubt they have chosen their sojourn there 
because of the facilities for fishing among the reefs.* 

Towards 10.45 a.m. we perceived the land of New Zealand, which 
we gradually approached by aid of a breeze from the N.N.W. The 
natives liad liglited a great fire on the most elevated of the hills that 
bordered the sea, and which advanced towards tlie North Cape. By 
5.30 we were not very far from tliat Cape, when two canoes were seen 
to approach from the shore. They were not long in reacliing us, and 
remained some time astern of us before daring to board us ; but after 
judging that our dispositions towards them were friendly, they 
approaclied with confidence ; besides, these Savages knew quite well 
that the Europeans which had visited their coasts had never been the 
first aggressors. They showed us at once some bundles of the fiax of 
New Zealand (phormium tenax) by shaking it to let us see its fine 
quality, and- offering it for excliange. It was with signs of great 
satisfaction that they received from us stuffs of various colors, and 
always with the greatest scrupulousness returned the price agreed on. 
They gave to iron a very great preference to ail other objects we 
offered tlieni. That metal is of so great value to this warrior people 
that tliey burst out in joyous shouts so soon as tliey found that we 
possessed the article, althougli we 4id not at first show but a small 
quantity of it, and at a distance. Nevertheless tliey recognised it 
perfectly by the sound made by two pieces struck one against anothei*. 

These people gave us in exchange for our things almost everything 
they had in their canoes ; we regarded it as a great mark of their 
confidence in us that they made no difficulty in depriving themselves 

♦ The Tlnee Kings were sometimes inhabited for years together by members 
of the Au-pouri tribe, when they had suffered defeat at the hands of the Rarawa, 
or other tribes living to the south of them. The middle island is not nearly so arid 
as the narrative makes out. In 1887, when Mr. Gheesemau, of the Auckland 
museum, and I visited the island we found quite a number of New Zealand trees, 
or shrubs, growing there, and apparently a large space sloping to the south had at 
one time been cultivated. There is water just below this spot, and it was just off 
there that Tasman attempted in 1642 to land and obtain water, but the attitude of 
the natives was so hostile that he decided not to attempt it, though much in want 
of water. There is a species of karamu growing there that has berries as large as a 
hazelnut, and another plant, the nearest habitat of which is China. The Maori 
name of the group is Manawa-tawhi. 



Digitized by 



Google 



The ¥isii of Dentnecasteaux to the North Cape. 119 

of their aims in our favour. The longest of their spears we got were 
not more than five metres long with a thickness of four centremetres — 
the smallest were half as long. They were all made of a single piece 
of very hard wood and perfectly polished. 

They gave us lines and fish-hooks, to some of which were attached 
feathers in place of bait that they use to attract the voracious fish. 
Many of these lines were very long with, at the extremity, a piece of 
hard serpentine which they attach to cause them to descend to great 
depths. We admired the fine polish they had given to the stones, 
which were of a spherical form, surmounted by a protuberance 
tlirougli which they had pierced a hole for the cord. It must be vei*y 
difficult for these Savages to pierce such hard stones, and no doubt it 
takes a long time, but they have plenty of leisure to devote to such 
work, for their wants are not many, and the sea supplies them 
abundantly with food. They sold us plenty of fish they had caught ^ 
there are such great quantities along the coast that during the short 
time we laid- to we saw many shoals of them on the surface of the sea^ 
agitating it over wide spaces, producing the effect of currents, and 
during calms they migiit be taken fur shallows. 

These Savages even deprived themselves of their clothing in i>rder 
to procure our objects in exchange. Some young men had ear-pendants 
made of serpentine of a very great hardness ; they were cut in ovals^ 
and most of them were about a " decimetre " (four indies) in length. 

The men had hanging on their breasts a species of human 
^^ cubitas " (P bone of forearm) at the end of a cord passed round their 
necks (see plate 25). They attach a great value to these ornaments. 

It is known that these people eat with avidity human flesh, and all 
that incites in them a similar idea causes them gi*eat pleasure. One of 
tlie sailors offered to one of them a knife, and, wishing to explain its 
use, he pretended to cut off one of his fingei*s, which he immediately 
carried to his niouth, pretending to eat it. Immediately the Savage, 
who had observed all the sailor's movements, expressed an extreme joy, 
and we saw him laugh for some time with his throat open and rubbing 
his hands. All of these people were very tall and veiy nmscular. They 
left us shortly after sunset. 

• At the same time a third canoe arrived from the nearest shore, 
manned by twelve islanders, who immediately demanded some hatchets 
in exchange for their things. After one of them had obtained one, 
another addressed us in a loud voice, shouting with all his power, " E 
toki " (an axe), and he ceased not until he got one. 

It was now night. The " Esperance " was too distant from our 
vessel to be seen, so we made flares in order to ascertain her position ; 
but we saw with surprise that the natives, far from showing any fear 
at the effect of gunpowder, continued their exchanges. It was more 
than an hour after dark that they paddled away to the coast. 



Digitized by 



Google 



120 JOURNAL OF THE POLY N ESI AM SOCIETY. 

While we lay -to the lead was cast several times; the soundings 
always showed a sandy bottom at sixty to eighty-three metres. 

23rd March. To the land breeze, which was felt feebly during the 
night, succeded at daylight a wind from the north-west. We were 
still very near the coast, and it would have been easy to have anchored 
in Lauristan Bay, but the sad event that had occurred to Captain 
Marion, and afterwards to Captain Fenieaux,*, determined the 
General to pass on. 

Nevertheless, I believe it my duty to express how important it 
would be to obtain in New Zealand the plant known under the name of 
phormium tenax (the flax of New Zealand) to transport it into Europe, 
where it would succeed perfectly. The firbe obtained from its leaves has 
a strengtli much superior to all other vegetable products which are 
employed in rope making; the cables made from it would resist every 
effort to break them. No one more than the Commandant of our 
expedition appreciated the utility of that plant for our marine. 
Nevertheless we continued our route towards the Friendly Islands by 
directing our course to the north-east." 

Tims ends Labilliardi^re's account of their visit to New, Zealand. 

On the 26th of the same month they discovered the rock — for it 
is really nothing more — which they called L'Esperance after the ship 
commanded by Huon Kermadec, whose name they also applied to the 
group, of which L'Esperance is the southern member. 

Both Dentrecasteaux and Kermadec died of scui-vey on the way 
home. It is strange to find these sailors bearing Military titles — 
General and Major— while Labilliardi^re calls himself ** Citizen," all, 
no doubt, due to the changes introduced by the French Revolution. 



♦ This refers to the massacre of Ferneaux's boat's crew in Qaeen Charlotte 
Sound no doubt. 



Digitized by 



Google 



NOTES AND QUERIES. 



[288] Malayo-Polynesian. 

In the Transactions of the Fiji Society for 1918, page 4, Mr. Coleman Wall 
refers to the Polynesians under the above title. In an interesting paper Mr. 8. 
W. Button, on page 18 of the same publication, refers to the Polynesians as 
Malayan. It is a pity these gentlemen iu both these interestuig papers continue 
to refer to a theory long since obsolete. It is now recognised that the Malays are 
a Mongolian race, while the Polynesians are a branch of the Caucasian. 

[289] The Pacific Story of Jonah and the Whale. 

In Mr. Coleman Wall's paper, referred to in the last notes, he gives the 
Fijian version of a tradition, somewhat analogous to that of the Biblical story of 
Jonah and the whale ; but in this case it was a woman who was swallowed by a big 
shark, and who eventually became the ancestress of a line of chiefs. The people of 
Nine Island have very nearly the same story, the incidents are very similar, except 
that the woman became an ancestress of some Tongan clan, for it was on Tonga 
that she landed and cut her way out of the big fish. So far as is known this story 
is not common to Polynesia, and the finding of it in Fiji and Niue, again shows 
that the latter people have a considerable element of the Melanesian in them, as 
insisted on in ** Niue and its People " — published by this Society. 



Digitized by 



Google 



PBOCEEDINGS. 

POLYNESIAN SOCIETY. 



A MsBTiNa of the Council was held at the Library, Hempton Room, on the 25th 
June, when current business including correspondence was dealt with. 

New members elected : — 

R. Clinton Hughes, New Plymouth. 

Dr, W. M. Thompson, M.A., M.B., B.C.L. 

The following corresponding members were also elected : — 

Thos. G. Thrum, Honcdulu, in recognition of his work in translation of 

the four volumes of Fornaiider's ** Hawaiian Antiquities.*' 
Stephen Savage, Rarotonga, in recognition of his translation of 

** Earotong^an Traditions.** 

Papers received :— 

Further Notes on the Heo. Rev. C. E. Fox, Solomon Island. 

Gilbert's account of Easter Island. H. D. Skinner. 

The Gt)ds of Maori Worship. Hare Hong^. 

Traditions and Legends of Murihiku. Part X. H. Beattie. 

Polynesian Linguistics. Part IV. Sydney H. Ray^ 

Traditions and History of Rarotonga. Parts VI. and VII. 

The death of our Corresponding Member, M. A. Leverd at Tahiti, from 
influenza, was reported, and the following resolution relating thereto was placed 
on the minutes : — ** That the Council has learned with great regret of the death 
of M. A. Leverd at Tahiti, and places on record its appreciation of If.. Leverd's 
contributions to Polynesian Ethnology — a subject in which he showed more than 
ordinary interest and ability with great promise for the future. And the Council 
further desires to extend to the relatives of the deceased its sincere sympathy in 
their loss.** 



Digitized by 



Google 




M. ARMAND LKVERD. 

Died at Tahiti, of influenza, 9th December, 1918. 



Digitized by 



Google X 



Digitized by 



Google 



THE LAND OF TARA 
AND THEY WHO SETTLED IT. 



THE STORY OF THE OCCUPATION OF TE WHANGA-NUI-A-TARA 

(the great harbour of TARA) or port NICHOLSON, 

BY THE MAORI. 



By Elsdon Best. 



PAET VIII. 

f Continued from page 96, Vol. XXVIII.J 
THE MAORI TEXT. 

TE HEKENGA MAI O NGATI-IRA. 

Na Rua-waliiiie Tawake i ki atu ki tona mokopuiia : — *'*Haere ki 
kaiiiga noho ai, ka watea lioki te wbeinia." Mo Waimata, mo 
Hikuwai, mo Tauwhai'e-parae, mo Huiarua, mo Te Ahi-kouka, ma 
Wai-iigaromia, koia nei aua wheiiua i ki atu ai a Tawake. 

Ka utua e Te Wha-kumu te kupu a tona tipuna ki aia ; — " Waiho 
o kainga hei liaehae i a koe, ina an e kaiiiga uei e te matao. Ka liaere 
tenei au ka whakamau atu ki te uru o te touga, ki te whare i maru ai 
au." Mo Tiitapora, i moe i a Rerekiokio, tona papa. 

Ko te putake tenei i lieke mai ai a Ngati-Ira i te pa tahuri i 
Pakauraiigi, e kiia ra ko Te Pueru-maku, uolio rawa mai ko Tapuwae- 
tabi-o-Rougokako i te pito ki te tonga o Whangara, i te taba rawbiti 
o Turanga-uui. I reira te pa ka mabia e Ngati-Ira, ka oti, ka nohoia 
taua pa e ratau. E rite te rabi o te wabi i uoboia e Ngati-Ira i roto i 
taua pa i Te Tapuwae ki te maara nei ; ka titiro atu matau ki taua 
maara, era pea e tae ki te wbitu eka te nui. E ki ana a Rihari i te 
taa 1837, e tu tonu ana nga awakari o taua pa i te taba tonu o te 
moan a nui. Ka hui a Ngati-Ira ki reira mahi ai i te kai, kumara kao, 
ika maroke, paua maroke, konra maroke, kao korau maori, arube 
parabou, arube kopuwai, ko nga arube pai tera, momona. Katabi ka 
wabi a Ngati-Ira, ko tetabi wabauga ka boki ano ki runga i te wbenua 
noboai; ki Tauwbare-parae, ki Haiarua, ki Waimata, ki Hituwai, 
ki Taumata-patiti, ki Anaura, ki Te Abikouka, ki Wai-ngaromia. 



Digitized by 



Google 



124 JOURNAL OF THE POLYNESIAN SOCIETY. 

Ko iiga taina o Taiie-ka-tohia, ara ko Rna-waliine me te tiiiigane, ine 
Tama-kauwae. 

Na, ko tetahi wahanga o Ngati-Ira ka ki e kore ratau e lioki atu 
Ano ki ena wlienua noho ai, ka liaere rataii ki te walii i u mai ai nga 
waka c nga kauinatua i te haerenga mai i Hawaiki ki Wliaiiga- 
paraoa; kola tera e iiolio mai ra a Te Tatana, a Tikitiki-rangi, me o 
ratau iwi o Ngati-Ira e nolio mai ra i Opotiki ra. 

Na, ka ki te wehenga ki a Te Wha-kumu, a ko ia me oiia hapu me 
liaere ki te pu o te tonga, ki Wairarapa. Koia teiiei te Ngati-Ira i a 
Taiie-katobia e nolio nei i Wairarapa nei ki a Te Miha-o-te-rangi, ki 
a Te Mauihera Ran gi-takai\v alio, ki a Tutapakilii-rangi ; kati nei aku 
« wliakahua ake i naia nei. Na, ko nga uri a Rua-wahine ko Te 
Aitanga-a-Maliaki, ko Te Aitanga-a-Haniti, taie atn ki a Te Wlianau- 
A-Rna i Tokomaru. Na, ko nga uri a Tama-kauwae koia tera a Ngati- 
Purou e pae mai ra i Tawhiti liuri noa ki roto o Waiapu tae noa ki 
Wliare-kaliika; kati mai i koiiei taku wliakaliuahua. He iwi nui 
tenei iwi a Ngati-Ira, nona teiiei wliakatauki : — " He pekeha ki te 
moan a, ko Ngati-Ira ki uta " ; he iwi toa lioki ki te pakanga. Kati 
taku whakamarama ake i tenei take ; ka pa lie tere koe ki te tulii, ko 
tenei ka taka te inarama i a taua e mahi ana kaore ano i paneke i a 
koe. 

Na, ka haere mai te ope a Te Wlia-kumu, ka tae mai ki Here- 
taunga, ka mahia tona pa, ko Nga Wliakatatara te pa, kei rawalii atu 
o Te Pa-wliakairo taua pa. I te tau 1863, i te tuwliera tonu nga 
awakari o taua pa. 

Ka kite nga iwi o Orotu, e, he iwi hou tenei kua hanga pa rawa ki 
runga i to ratau whenua, a katahi ka whakawhaiti Tini o Orotu, ara a 
Rangitane, ka whaiti ki roto i Te Pnketapu pa, i rawahi mai o Omahu 
i Heretaungara. Katahi ka mea a Paewhenua, a Te Hau-te-rangi, 
a Te Kowhaiwhai, nga rangatira a aua iwi, kia kotahi te matuae tnkn 
ma roto i Tutaekuri, ko tetahi me tuku ma te hiwi iTe Tauwhareheke 
iho ai ki Nga Wliakatatara pa o Ngati-Ira taupoki ilio ai i te pa. 
Na, tetahi o nga matua me tika atu ma te parae ki Te Wai-o-hiki tau 
ai, kia marama ai te titiro atu ki te putanga mai o te matua i roto o 
Tutaekuri, o runga hoki o Tauwhare, hei poa hoki i a Ngati-Ira kia 
haere mai ai ki walio o tona pa, kia riro ai te pa Nga Wliakatatara i 
te matua e heke iho ana i Tauwhare, kia watea ai te matua i roto o 
Tutaekuri hei awhina i a ratau ka whakaeke ra ki Te Wai-o-hiki pae 
ai. Koia tenei nga whakaaro i roto i a Rangi-tane, i a Ngati-Awa, i 
a Ngati-Mahanga. 

Na, i te poka whakatata ki te whakaeroero nga whetu i te ata 
hapara, ka wehewehe nga niatna e torn nei, he ran topu ki te matua 
kotahi. Na, ka hapara te ata whakaao marama, ka puta atu h Te 
Ahipara, a Te Horipu ki waho noho ai, ka kitea atu te matua tangata 
« haere mai ana i te mania, ko te kirikau anake. Ka hoki ki roto i te 



Digitized by 



Google 



The Land of Tara. 125 

pa, ka karaiiga: — " Ko Tu-mataenga ! Ko Tu-matauenga ! " Ka 
karanga tetahi o ana taiigata : — " Ko te whakaanki ! Ko te whaka- 
ariki ! " Ka karanga tetalii : — " Kei te mania." 

Ka puta niai a Te Wha-kumu ki waho o te pa niatakitaki atu ai. 
Kh mea nga toa o te pa o Ngati-Ira kia tikina kia whawhaitia taua 
ope. Ka mea a Te Wha-kumu: — "Kia man, waiho kia ukiuki te 
whenua, akuanei e whakaputa ana ki Te Puketapu, kua kore he 
tangata o roto, he wahine, he pangore anake kei roto. Kia wlia ran e 
kawhaki ki reira, ko te tokomaha e noho hei ora mo te pa nei. Kaua 
rawa hei puta kei waho, waiho kia awhi ana i te pa nei. Ko tena e 
haere mai ra he alii hunnhunu, ana ano te matua kei te ngaro i a tatau. 
E hara tena, he patoi kia puta atu tatau ki waho, ka manukawhaki ai,^ 
ko reira te matua huaki ai. Koi pohehe ki tena ahua arataki matua." 

Ka noho katoa a Ngati-Ira, taiie, wahine, tamariki, i roto i te pa 
noho ai< Ka piki a Te Wha-kumu ki runga i te puhara noho ai,^ 
mataki ai. Ka mea ia ki ona toa: — ** Whakaarahia he puhara moku,. 
kia kotahi ki te taiia ki uta, kia kotahi ki te taha ki te awa nei, kia 
tiketike," t 

Kua takoto nga rakau, ka whakaarahia, ka ara ki runga ana 
pnhdra e rua. Ka eke a Te Wlia-kumu me nga toa tokorima ki te 
puhara o uta ; ko te puhara i tai ka eke a Te Whanonga me nga toa 
tokorima hoki, me a ratau tokotoko, me a ratau manuka kanoi, me a 
ratau pukoro kowhatu hei whakaruru ki te taua nei. 

Na, i penei te ahua o te pa nei me te tu a nga puhara nei. Ko nga 
waha ngutu tera i raro iho o nga puhara e rua. Ko te tu a nga 
puhara e rua nei he tiaki i aua waha ngutu e rua, koi uru te tangata 
ki roto i te pa nei. Tuarua o nga take he whakaatu i te ahua o te 
mahi a nga taua awhi i te pa, whakaeke ranei, tukituki ranei i te 
parepare o waho, ki nga toa o roto i te pa. Na, ko te kuwaha o te 
parepare o roto, kotahi tonu te kuwalia e tomo atu ai i te parepare o 
waho ki roto i te parepare o roto. Ko taua waha ngutu he mea hou 
atu i raro i te whenua, ka puta ai ki roto o te parepare tuarua. Na,^ 
pera ano te waha ngutu o te parepare tnatoru, he mea liou ma roto i 
te rua e uru ai ki roto ; ko nga parepare e torn katoa. Na, e kite ana 
koutou i te parepare kokoti o roto rawa; ko tera he rohe mai i nga 
wahine, i nga tamariki, i nga koroua ki te wahi whaiti noho ai. Ko 
te waha ngutu kei roto i te whenua ka puta atu ki roto ano i te pare- 
pare paku na e puta ai ki roto i te wahi o roto rawa me hou ano ma 
roto i te whenua i te waha ngutu. Na ko tetahi wahanga o to roto he 
marae tera no nga tane anake e takatu ai ki te tiaki i nga parepare o 
te pa nei, ara ko te wahi e tenei (?). Ko te parepare o waho rawa e kiia 
ana e torn whanganga te tike tike ; ko te awakari o waho a taua pare- 
pare e wha whanganga te whanui me te hohonu o te awakari. Na, 
ko nga parepare katoa o roto mai i to waho, kaore he awakari, engari 
e rua whanganga te tiketike ake i te papa o te pa <iei, e wha te papa 



Digitized by 



Google 



126 JOURNAL OF THE POLYNESIAN SOC/ETY, 

whaiiga (?), e rua a runga, haere toiiu ai iiga taiigata i ruuga, ara nga 
toa. Ko te parepare o waho rawa kotahi te wliaiigaiiga te matara o 
tetahi i tetahi, ara o to waho rawa i to roto niai, i peratia ai ino te uru 
te taua ki roto kaore e tika toiia wero i toiia tokotoko, huata ranei, i 
te apiapi o aua parepare, a kei iniiiga ra hoki nga toa i te parepare o 
roto e haereere aiia, e werowero ana i nga tangata e uru ana niai, ina 
uru ki roto i taua pa. Kati aku whakamarama i te ahua o te pa nei. 

Na, ka roa e noho ana ka wbakaputa mai te wahanga o te taua i 
ma roto mai i Tutaekuri awa. Na, e lieke iho ana hoki te matua i 
heke iho i runga i te hiwi o Tauwhare, kua whakawhiti mai hoki te 
wehenga i Te Wai-o-hiki ra, ka karapotia te pa nei ka tahuri ki te 
whakauru ki roto i te pa nei. I te nui o te mahi a nga tangata o te 
taua ki te whakauru, kore rawa nei i taea i te wehi i nga pnhara e rua 
nei, a tokotoru rawa nga tangata i mate i te werohanga ki te tokotoko, 
ki te huata hoki, ko Te Horeta, ko Hauparua, ko Te Iwi-katea, koia 
nei nga mea i mate o te taua i tenei ra. Ka po hoki, ka heke te taua 
ki te talia o te awa noho ai i taua po nei. 

Na, ka tonoa e Te Wha-kuum nga tangata e rua rau topu ki Te 
Puketapu pa i runga i te puke. Ka taliuri taua pa, ka riro herehere 
mai nga wahine, nga tamaiiki, nga kaumatua. Ka mate i reira a 
Koura, a Te Awapara, a Te Kirinia, q. Poupou, a Tangi-akau, me era 
atu e maha, era pea e tae ki te whitu tekau, neke ake ranei. 

Na, ka hoki nei te iwi nei, tera tetahi wahine kua puta i te tana nei, 
kaore i kitea, i roto i te rna kuniara e moe ana. Ka warea te taua ki 
te hopuhopu i nga tangata o te pa nei, ka heke te wahine i te puke i 
runga ake i te piriti o Omaliu, ka tae ki te tana ka ki atu kua tahuri a 
Te Puketapu. Ka maharatia he taua na Ngati-Whiti-kaupeka no 
Patea, no Rangitikei ranei. Ka wliati te taua nei i konei, ka tae ake 
nga tutai ki a Te Wha-kumu, ka mea : — ** Kua tae niai te karere o te 
taua ki Te Puketapu, kua tahuri te pa. Kati, kai te whati te taua nei 
ki reira." 

Ka mea a Te Wha-kumu ki a Te Okooko, ki a Kokau : — " Haere 
korua, kia wawe koinia ki te taha ki mua o te taua i whati nei, 
whakatitaha ai i te ara hokinga mai ma uta ma runga hiwi, kia hori 
atu te taua ka heke iho ai ki te ara nui ; ka whai atu hoki matau." 

. E korero ana ka kitea atu ka koori (? koiri) te au o te ahi o te weranga 
o Te Puketapu i te ahi, na Te Nan ara i tahu. Katahi ano ka rere nga 
toa nei, ka mea a Te Wha-kumu kia rua rau topu e haere ki te whai i 
te iwi nei. Ka whaia i te po, ka tae atu ki Te Awatapu, kua kino te 
whati a te taua nei, kua marara noa atu te haere i te whawhai, kia 
wawe te tae. Ka timata te karapoti o te rua rau topu ; ka patua 
haeretia te iwi nei ; koia tenei parekura a Marae-kakalio. No te liora 
haere i te parae o te tangata, kaore rawa i taea taua pa e te taua te pa 
horo. 



Digitized by 



Google 



The Land of Tara. 127 

Na, ka roa, ka pae he kai ma ratau, katahi ka liaere ki Wairarapa. 
Kaore i ea tenei inatenga i a Ngati-Awa, i a Ngati-Mabanga, i a 
Kangitane hoki. 

Ka tae te taua nei ki Wbareama, a Ngati-Ira, kaore i homai he 
waka hei whitinga i te awa o Whareama. Ka iioho i Wai-mimiha, i 
te ngutuawa o Whareama ; katahi ka wliakaturia te haka e te wahine, 
iiawai ra ka eke inai i ruiiga i uga waka kia tata mai ratau ki te taha 
matakitaki ai. Ka neke te liaka a nga wahine o Ngati-Ira ki uta, 
katalii ka u mai nga tangata ki uta matakitaki ai ki a Ngati-Ira e 
haka ana. Katahi ka huakina mai e te taua ma roto i te tahataha o 
te awa, ka mau e wha nga waka. Ka kokirikiri nga tangata ki roto i 
te wai, ka kau ki te taha tonga o te awa o Whareama. Katahi ka 
whakawhiti a Ngati-Ira i Wliareama, kotahi mano topu taua taua a 
Te Wha-kumu ; koia te vvhakatauki na : — " Tena, tera a Ngati-Ira te 
haere na i uta me te mea tera he tere pekeha i te moaiia." Mo te nui 
o te tangata o taua ope. 

Na, tera kua tae te rongo o te taua nei ki nga wahi katoa, me te 
nni o te wehi o te tangata. Kua omaoma nga tangata noho tahora, 
me nga hapu me nga iwi ki roto i nga motu o uta o te tuawhenua, 
i te nui o te wehi. Ka whakaturia te pa mo te whawhai ki a Te 
Wha-kumu me tona iwi. Ko te Pa o Bakai-tauheke tetahi, kei uta o 
Wliareama ; i tu ai ki reira, i mahai'atia era taua ope e tika atu ma te 
ara e tika ai ma uta ki Wairarapa ; ko Ihutu te pikitanga tiketike o 
taua ara. Kia tae atu ki reira ka whawhai ai a Rangitane, a Ngati- 
Whutumamoe, koia nei nga iwi kua hui ki roto o taua pa. Kotahi o 
nga pa, ko Nga Wahine-potae, kei runga i te maunga e tu ana, kei te 
taha rawhiti o te awa o Mangapakia ; tenei pa i aua iwi ano. Kaore i 
mamao te Pa o Rakai-tauheke i Nga Wahine-potae. 

Na, ka mahia tetahi pa ki te taha tonga o te ngutuawa o Whare- 
ama, ko Oruhi te ingoa, lie karapuke nei te tunga o taua pa ; no 
Rangitane ano taua pa. Ka mahia tetahi pa ki Tupapaku-rua, ki uta, 
ko Take-whenua te ingoa, he pa nui kei te huanui e ahu mai ai ma nta 
ki Maungarake nei. Engan he pa tahito enei pa e inia, a Oruhi, a 
Take-whenua; ko te Pa o Rakai-tauheke, he pa hou tenei ; ko Nga 
Wahine-potae he pa tahito tenei no Ngati-Wairehu, no Ngati-Takawa. 
Enei hapu, iwi hoki, no Wliata, koia te rangatira o enei iwi e rua, e 
aranga Te Kai-hinaki-a-Whata i Te Waipukurau ra o Heretaunga ra. 
Kati, ka niarama mai koe ki enei take. 

Na, ka kite koe kua haere te ope taua a Te Wha-kumu ma Whare- 
ama ki te ngutuawa,; ka wliiti katoa te taua nei ki te taha tonga o te 
awa nei.. Katahi ka whakaeketia a Oruhi, kaore i roa ka horo taua 
pa, ka horo i te po, ka mau a Te Poki, a Kaikore, a Te Whatu-rakau, 
a Te Hau-taruke, koia tenei nga rangatira o Rangitane i mate ki 
reira. Ka horo atu etahi ki roto i Take-whenua i Tupapaku-rua, i 
kiia ake ra e au. . 



Digitized by 



Google 



128 JOURNAL OF THE POLYNESIAN SOCIETY. 

Ka roa e iiolio aim ki te mahi kai moan a ma ratau taua ope o 
Ngati-Ira, ka pae te kai moana, te aruhe, te mahi. Ka mea etahi o 
nga herehere ki a Ngati-Ira : — " Kei uta o Whareama iiga pa e rua e 
tatari ana mai ki a koutou i te nui o to koutou roiigo toa e liau mai 
nei i te vvheiiua; ko te Pa o Rakai-tauheke, ko nga Wahiiie-potae, 
ka oti tera te kowbiri ko ia toa anake o Rangitaiie e noho mai nei. 
Kati, kotalii o nga pa tauwhanga i a koutou ko Take-when na, kei uta 
o te pae maunga nei, kei konei anake te toa o tenei takutai moana e 
whanga ana mai ki a koe." 

Ka mea a Te Honoiti ki a Te Wha-kumu : — " Kua pae te ora o te 
tangata ; pewhea tatau ? " 

Ka mea atu a Te Wlia-kumu ki a Te Honoiti me Ngati-Ira katoa : — 
*' Tenei te whakaaroaro ake nei te ngakau ki nga rongo korero a te 
herehere nei, kei nga huarahi e tauwhanga mai ana nga p.a tuwatawata 
tutaki i a tatau, whakatau hoki i to koutou rongo toa o te riu o 
Heretaunga, haere mai nei ki tenei. Kati, ki te titalia tatau i te Pa 
o Rakai-taulieke me Nga Wahine-potae ka wliai kupu a Rangitane, a 
Ngati-Mamoe — Miei aha nana ra i titaha penei takoto wharoro ana 
ratau i konei.' Koia tenei te ahua o te kupu ma ratou ki a tatau, 
koia ahau i whakaaro ai me wahi ki a koutou kia wha rau topu ki 
Nga Wahine-potae, kia wha rau topu ki te Pa o Rakai-tauheke, 
whakaeke ana telahi me tetahi. I te mea kua tae te rongo kua titaha 
tatau i aua pa e rua, kua marara te noho a te tangata o roto, ka 
whakaeke ngawari noa atu te taua. Kia wha topu e waiho hei tiaki i 
nga wahine, i nga tamariki nei i konei, ko tatau e haere ki era pa e 
rua." 

Ka wehea te haere a nga matua nei, ko tetahi ka ahu ma te taha 
tonga o Whareama, ko tetahi ka haere ma te taha rawhiti marangai o 
Whareama, a ka tata atu ki te Pa o Rakai-tauheke, ka noho i reira 
kia tae tetahi matua. Ka tae te matua ma Mangapakia, he awa tera^ 
tae atu ki Te Papa-kowhai i raro iho o te pikitanga o Ihupiri, ka noho 
kia awatea. Ka huaki i te po, ka whakaawhitia nga pa nei, ka tatata 
ki te puao nui ka kitea tetahi tangata e haere ana i vvaho mai o te pa, 
me te wahine ; ka hopukia, ka man, ko Kapukapu me Hine-whiri aua 
tangata, no roto i te Pa o Rakai-tauheke. Ka uia atu : — " Kei te aha 
a roto i te pa nei ? " 

Ka kiia mai : — " Kua hokihoki te tangata ki uta whenua ki nga 
kainga noho tahora, ki nga wahine, ki nga tamariki." 

Ka uia atu ano : — " Kei te pewliea a Nga Wahine-potae ? " Ka 
utua mai : — " Kua pera ano te ahua o te tangata." E mahara ana a 
Kapukapu raua ko Hine-whiri he taua whakauru tera no Rangitane 
ake ano. Katahi ka ni atu a Kapukapu : — " Kua hori te ope taua a 
Te Wha-kumu me tona iwi, a Ngati-Ira, ma Tupapaku-rua ? " 

Ka mea atu te tangata o te taua : — " Koia ra tenei te taua a Te 
Wha-kumu e ki mai na koe." 



Digitized by 



Google 



The Land of Tata. 129 

Ka kanakaiwi nga wliatii kia oma ia ; ka inea atu a Te Honoiti : — 
" E ! kaua hei oma, koi patua koe, ata nolio kia ora ai koe." 

Ka whakaura te Iiaeata o te ata ka whakaeke na te Pa o Rakai- 
tauheke, ka mate a Rakai-tauheke i konei. E kiia ana he iiui ano te 
tangata i mate i reira, he tane anake te iminga. Ka mauria te upoko 
o Rakai-tauheke ; te take i mauria ai he pai uo te ahua o taua tatigata, 
e kiia ana he tino taugata pai rawa atu te ahua; ka mauria te iipoko 
hei whakaatu ki te iwi o Ngati-Ira. . 

E tahuri ana tenei pa ka wera mai hoki a Nga Wahine-potae i te 
ahi. Pera ano kaore i maha nga tangata o roto ; i roto, ki roto i te 
awa o Mangpakia te nuinga o te tangata, i reira e paepae ake ana te 
taua a Ngati-Ira, i wehea ko etahi hei whakaeke i te pa, ko etahi hei 
te taha ki te awa noho ake ai. 

Ka hoki nga matua taua nei, ka tae ki Oruhi, ki to ratau nuinga. 
Ka mea atu a Te Wha-kumn : — "Kati. Tuku atu aKapukapu me nga 
hereliere mai o Nga Wahine-potae, 6 te Pa o Rakai-tauheke kia haere." 

Ka tukua katoatia nga herehere kia haere. Ka mea atu a Te 
Wha-kumu ki a Kapukapu me era atu o nga herehere : — "flaer^! ki 
atu ki a Rangitane me era atu iwi, whakawatea ki tahaki i te ara 
moku i Tupapaku-riia. E haere ana au ki Potaka-kura-tawhiti, ki Te 
Wharaunga-o-Kena, ki aku matua ki a Te Whakamana, ki a Te 
Rerewa. Kaore au i haramai ki te patu tangata; i mate ai te tangata 
i au e haere nei, he whakawatea naku i taku huarahi ; i peka ai au ki 
uta o Whareama nei, he wawao naku mo te kupu he wehi i titaha ai 
taku ope ma Tupapaku-rua. Mo tena whakaaro anake i peka atu ai au 
ki te whakatau i Rakai-tauheke me nga Wahine-potae, Haere ! 
koi tahuri mai o koutoii kanohi ki muri nei ; kia maro te haere ki 
tahaki o te huarahi." 

Ka haere a Kapukapu me on a hoa e rua ran tuma, te tane, te 
wahine, te tamariki. Ka nui te koa o nga herehere mo te tukunga i a 
ratau kia hokihoki ki o ratau kainga. Ka mea a Kapukapu, a Te 
Whao, nga tangata o nga herehere ra : — "Haere! ma maua e haere 
ake ki Take-whenua nei whakaatu ai i to kupu. Ki te whakaae mai, 
tena maua e hoki nmi ki a koe ; ki te kore e whakaae mai 
ka hoki matau ki Puketoi kei uta o Whareama, o Mataikona, o 
Owahanga hoki," he maunga tera wahi a Puketoi. 

Ka po rua e whanga ana te ope haere a Ngati-Ira ki a Kapukapu 
raua ko Te Whao, kua kore e tae mai. Ka mea a Te Wha-kumu ki a 
Te Honoiti : — ** Whakatika tatau ka haere." 

Katalii ka haere, ka tae ki waho mai o Take-whenua, ka pa mai te 
reo o te tangata o roto o te pa: — ** Aue ki au ! E koro ma, e! Te 
takoto kino mai ra i ro o Whareama . . e . . i." 

Koia ra te niaioha a nga wahine o te tangata whenua. Ko te ahua 
o taua maioha e niaioha ana mo nga tangata i mate ki nga pa i kiia 
ake nei o roto o Whareama. Engari kaore ano i puta atu te taua nei 



Digitized by 



Google 



130 JOURNAL OF THE POLYMESIAM SOCIETY. 

ki waho o te putaanga; i ki atu a TeWha-kumu : — " Ko te liaere, me 
tutira te haere, kia kotahi ran topu te inatiia ki mua, me tane anake ; 
kia kotahi ran topu te waliine e maka ki waenganui. Ka aua atu tera, 
ka tukna ake kia kotahi rau topu te tane ; ka aua ake tenei ka tuku 
ake kia kotahi rau topu ano te wahine ; me pena tonu, a inutu uoa. 
Kia wha rau topu te matua tane ki muri hei tutaki mai i muri." 

Koia tenei nga tohutohu a Te Wha-kumn ki tona iwi ki a Ngati- 
Ira; na, ka pera te haere a te iwinei. 

Ka whanga nga tangata o roto o te pa ra kia kitea te hiku o to ope 
nei. Ka pena tonu te liaere o te tangata, a ahiahi noa e haere ana. Ka 
mea nga tangata o roto i Take-whenua : — " Ka tika ano te toa o te iwi 
nei e hau mai nei te rongo ki a tatau, ina te ahua me he nru ngahere 
tera te nui o tenei ope." Ka tau te wehi ki nga tangata o roto i Take- 
whenua, kaore rawa i korikori te tane, ko te wahine anake e maioha 
mai ana ki o ratau tupapaku i mate ra i Oruhi, i te Pa o Eakai- 
tauheke, i Nga Wahine-potae. 

Ka tae te ope nei ki Wainuioru, ka tae atu a Kapukapu raua ko Te 
Whao ki te puni o Ngati-Ira, ka korerotia kaore i whakaaetia te take 
o ta raua haere mai, kia kaua hei whawhai : — *^ Kotahi tonu te 
whakaaro o te tane, o te wahine, ko te puta tonu mai ki waho ki te 
whawhai ki a koe, e Ngati-Ira ! Na te kitenga i te ahua o te rere a te 
tangata, a ahiahi noa, kaore e mataki te haere o te tangata, katahi ano 
ka mate haere te whakaaro o nga toa whakaputa o roto o Take- 
whenua. Kati, kua mutu, haere noa atu koe, e tokoto nei ki te liu o 
Wairavapa, kaore he aha o mua i a koe. Kei Potaka-kura-tawhiti 
anake te tangata, i to rongo, nana i whakawhaiti te noho a te tangata 
ki reira. E tae e koe ki Maunga-rake, e nolio i kona; tukua he 
tangata man ki o matua i roto o Potaka, kia wawe ai ratau te mohio 
mai ko koe." 

Ka mea atu a Te Wha-kumu : — ** E pai ana." Ka mea atu ano a 
Te Wha-kumu : — " Ko taku whakaaro me noho korua, ko korua tonu 
he karere maku ki roto o Potaka." 

Ka whakfciae a Kapukapu, a Te Whao. Ka tae te matua nei ki 
runga o Maungarake, ka marania te kanohi ki roto o Wairarapa, mai i 
te moana tae noa ki te kauru. Ka noho te ope nei i reira, ka tukua a 
Kapukapu raua ko Te Whao kia haere. Ka tae a Kapukapu ma ki 
Potaka-kura-tawhiti, ka uia mai e Te Whakamana, e Te Rerewa, me 
o raua iwi : — " I nu tai ?" Ka mea a Kapukapu: — "Ko Te Wha-kumu I 
Ko Te Wha-kunm te ope haere nei ! " 

Ka mea ano a Te Whakamana : — " E Kapu ! Hokia ano ! " Ka 
mea ano a Kapukapu: — "Ko Te Wha-kumu! Ko Te Wha-kumu a 
Tu-tapora te ope haere o Ira." 

Ka tukua mai nga taina o Te Whakamana me to ratau tuahine ki 
te kawe kai mai ma Te Wha-kumu, hokorima topu te ope pikau kai 
mai ma te ope haere mai o Te Wha-kumu. I konei ka tae mai ki 



Digitized by 



Google 



The Land of Tara. 131 

Mauiigarake nei te ope mai o Potaka-kura-tawhiti, koia nei nga mea i 
tonoa mai, haunga nga kai pikau kai. He kao korau, he kao kamara, 
he piharau, he tuna hao maroke, he ika paka, he koura mahiti, he 
paua maroke, he pepe whinau, he pepe iiianga, he taha buahua, koia 
iiei nga kai. 

I mnri i iho i tenei ka ki atu a Hine-tuwawe, tuahine o Te 
Whakamana, ki a Te Wha-kumu : — '^ E tama ! Me haere tatau ki te 
marae i o matua, o o tipuna, o o tain a, kia wawe te ta o ratau ngakau 
ki a koe, e honiai nei to rongo me he mura ahi tera e taoro ana to mahi 
ki te patu haere mai i a Rangitane, i a Ngati-Awa, i a Ngati- 
Mahanga, haere mai nei koe i te takutai nei kaore he morehu e tu ana 
ki to aroajo. Kati ra te patu i te taiigata, kei hea he whakaruru hau 
mou i patua ai koe." 

Ka mea atu a Te Wha-kumu : — ** E hine ! He tika to kupu, ka 
mutu taku, he karo patu, me i kore hoki au e karo i te patu moku, & 
kore au e kite i a koe." 

Katahi ka whakaae atu a Te Wha-kumu kia haere ratau, ka mea: — 
'' Kg te rakau nei he rata tona ingoa o mua, me ki i naia nei ko Te- 
Eata o Te Wha-kumu. '^ 

Ka tae ki Potaka, ka moe a Te Wha-kumu i a Hine-ipurangi,. 
tetahi teilei o nga tino wahine rangatira o Wairarapa nei o aua ra. 

TE TANGI A NUKU I TE MATENGA I PEEHIKATIA, MO TE 
OHONGAITUA, MO TE RANGI-TAKUARIKI. 

** Haere atu ra, e tama ma e ! 

I te mata o te rakau a Tu-matauenga 

I patua ai Kaupeka i roto o Kauwhata-roa 

Ka tangohia te manawa, ka poia ki a Aitupawa 

Ki a Eehua, ki a Tahurangi 

I te mata takitaki i tupea ai a Rangi 

Ki te poho o Hangi-tamaku i Tahuaroa 

I hikaia ai e Tupai, e Tama-kaka 

Ki te alii tapu na Kangi-nui 

I takaliia ki Tauru-rangi-ataniai 

Ka tu tona ahi, koia te ahi tapu 

Koia te ahi toro, koia te ahi tipua 

Ka puta ki te hou matapu 

Ka ea ki te ao nei — e tama ma, e ! 

Haere ra, e tama ma, e ! 

I te ara ka takoto i Taheke-roa 

Kia karangatia mai koutou ki te Muri ki te Waihou 

I to koutou tipuna, i a Buaumoko 

E whakangaoko ra i Earohenga 

Ka puta te hu ki te taiao 



Digitized by. 



Google 



132 JOURMAL OF THE POLYNESIAN SOCIETY, 

Koia Hine-puia i Hawaiki 

E tahi noa mai ra i te kauhika 

Ki waho i te moaua 

Ka tere Hine-uku. ka tere Hine-one 

Ka tere Para-whenua-inea ki a Hine-moana 

E tu mai ra i Tahora-nui-atea 

Ka whakapae ki uta ra, koia Hine-tapatu-rangi 

E haere atu na korua 

E tama ma . . e . . i." 

[The allusion above to volcanic action in Polynesia in past times is 
interesting. Hine-puia is the personified form of volcanic activity ; 
Hine-uku and Hine-one, representing earth and land, together with 
Para-whenua-mea, representing water, as streams, etc., move into or 
are engulfed by Hine-moana, personified form of the ocean. 
Tahora-nui-atea, Mahora-nui-atea and Marae-nui-atea, are all ex- 
pressions denoting vast ocean expanses, the waste of waters, sometimes 
alluded to as the marae or plaza of Hine-moana. 

The above allegory reminds one of a tradition preserved by the 
Takitumu folk of the submergence of certain lands in Polynesia in 
former times, caused or accompanied by a tremendous eruption that 
destroyed a mountain named Maunga-nui, in which catastrophe whole 
tribes perished.] 

HE WAIATA NA NUKU. 

Mo te korenga o nga tangata o Wai-rarapa e hoki mai ki te 
whakatau ki te whawhai ki a Ngati-Awa, ki a Ngati-Mutunga, ki a 
Ngati-Raukawa, ki a Ngati-Toa. 

" Nei ka noho i te ra o te warn 

Ka haramai e tamaroto ka pupuke ake 

Me ko Eua i te pukenga, ko Rua i te wananga . . e 

Rua i te rururu, e, ko Rua i te wetewete, e 

Ko Rua i te horaliora ki tukemata rau, e 

Waiho ra, e, me ata kaupeehi iho, e 

Kia ata tukutuku ra, e, i te aliorangi, e 

Me ui pea e au ki te makau tangata o te waotu, e 

E hara tenei kei te pokiki he paewai, e 

Me ui pea e au ki te mata ngaro i a Rangi, e 

Kaore te ki mai te waha 

Ka wlianatu tenei an, ka whetoki haere ki tai ra 

Me ko te punga i te toroa a punga 

Ka whai i te pua tuhi no Tawhiti-nui 

No Tawhiti-pa-mamao, e 

Kia whitirere ake au me ko te ata i marama 



Digitized by 



Google 



The Land of Tara. 135 

Marama te ata i Hotu-nuku, e 
Ko Tane te waiora . . e . . i." 

HE WAIATA TANGI MO NUKU. 

Na Te Whare-pouri i Nukutaurua. 

** Tera Tariao ka kokiri kai runga 

Ko te rite i ahau e vvliakawhetu nei 

Wairua i tahakura nou nei, e Nuku 

Kei te whakaara koe i taku nei moe 

Kia hua ake ai au ko to tinana tonu 

Me he wai wharawhara te tuturu o te roimata i aku kamo 

E tangi, e manu, kia mohio roto 

Ma te hau tonga au e wliiu 

Nga puke iri mai o Rangitoto i waho ra 

Kia whai atu au i to tira 

Ka wehe rawa ia koe i ahau 

Tera pea koe ka iria he inaunga a tai 

E horo e manu kahu i raro ra 

ifi nga puhi raia ki Wainuku 

Maruao ki Marianuku * 

Te huri rawa mai to wairua ora 

Ki ahau i kouei . . e . . i.'* 

HE WAIATA. 

Ka mau a Eipeka Te Kakapi, tamahine a Te Whare-pouri, me 
nga matua, ka tae ki Nga Umu-tawa, ka poroporoaki tera, koia t^nei 
tona tau : — 

** E tama ina, e ! Teuei au te rapa noa nei, 

Te hahau noa nei 

I te mate waiora na Paikea 

Kia ea ake ana ko Hikurangi nga morehu, e 

Tenei ia ka pahunu te tangata, ka pahunu te whenua 

Ka rapa te vvaewae i te manawa ka nguha 

Ki te takapau whare ki te whenua 

Ka hurihia ake nei ki muri 

He hau taua e topetope mai nei i te whenua, e 

Tenei he korouga ka tu ki roto o Makauri 

I whiua mai ra ki uta hei tohu mo Mahaki 

Kia kata noa mai te kikitara 

Kotipatipa kohureliure titipounamu 

E tangi haere ana ki tona whenua 

Ka tipuria nei e te maheuheu 

Tangi kau ana te mapu . . e . . i." 



Digitized by 



Google 



HISTORY AND TRADITIONS OF RAROTONGA. 



By Te Akiki-tara-abe. 



PAET VI. 

Translated by 8. Percy Smith. 



ABOUT TANGIIA AND TE NGA-TAITO-ARIKI. 

[In the following part (which is a continuation of Part IV.) the 
Sage, while professing to give the hjstory of the celebrated 
Earotongan chief, Tangiia-nui, also in reality gives a brief sketcli of 
the history of the people right away from the first ancestor known to 
them down to the same Tangiia, who settled in Rarotonga in the 
thirteenth century. Tangiia 's adventurous voyages, his wars, and 
his loves will form anotlier part of these papers. 

As in all these old Polynesian legends we are carried back to that 
stage of development in human progress, when it was the common 
belief that the gods took part in tlie affairs of mankind, a belief by 
no means exclusively Polynesian. 

The scene of most of the following story is laid in Savai^i and 
*Upolu of the Samoan Group, the former of which islands is known 
to the Earotongans as Avaiki, wliile Avaiki-raro is a general name 
for the Samoan, Fiji, and other islands in their neighbourhood. In 
«ome of the proper names it is difficult to understand whether the 
Sage refers to gods or men, for they often have identical names. The 
story of the E.uru (White Heron) and the Sea-snake is also to be 
seen in our ** Rarotonga Records,*' and the two stories whilst agreeing 
in the main, should be read together, for each contains detail not 
shown in the other. Some remarks on the genealogies are referred to 
in the general introduction to. this series of papers.] 



Digitized by 



Google 



Hisiory and Traditions of Raroionga. 



135 



This is a word about Te Nga-taito-ariki, who was a son of Te Tumu 
(from whom descend the following generations of men) : — 



1 95 Te Tumu 

Te Nga-taito-ariki 

Mna 

Eauga 

Maina 
90 Maeata 

Makaro 

Te Marama 

Te Eang^-ki-te-ao 

Tangpata-kato 
85 Te Atu-tangang^ 

Te Atu-te-ngangpata 

Te Atu-te-ki 

Te Atu 

Tiki 
80 Taito-rang^-ngfung^ru 

Taito-rangi-ioio 

Taito-kum-angiangi 

Taito-kuru-ma-rakamea 

Vaitakere 2 
75 Te Tarava 



74 



70 



65 



60 



58 



Te Tarava-enua 

Te Rua-enua 

Te Rna-mata-iko 

Te Punnpunu. Te Utarei (e 

uaDga ika ia. Anan akera 

ta Rua-roata-iko). 
Ara-kapua 
Te Ara-o-nga-atua 
Toi 
Pniig^verevere (ka ngae ki te 

Arn-o-nga-atiia). 
Mann 

Manu-kaiaia 
Manu-kavakevake 
Ore 

Turanga ma Te Rnru 
Rongo-rua 
Rira (or Rina) 
Te Irapanga 3 
Tu-tarangi 
Etoi 



TU-TARANGI AND HIS WARS. 

254. Tu-tarangi (generation No. 59 above) caused a great war, 
the reason of whicli was as follows : He owned certain birds [probably 
trained sea-gulls] named Aroa-uta and Aroa-tai ; they were trained 
birds that did work for him, they obtained food, and fished for him. 
On a certain occasion, Tane-au-vaka sent a messenger to Tu-tarangi 
asking for the use of his birds, but Tu-tarangi would not consent at 
first, but in consequence of frequent applications he at last allowed 
the bird that lived ashore (Aroa-uta) to be sent to Tane-au-vaka. 
But the bird would not act (for its new master) so it was killed. 
Tu-tarangi was then applied to for the other bird, and he finally 
consented to lend Aroa-tai to Tane-au-vaka. This time the bird did 
his work and caught fish ; but it was not treated properly, it was not 
fed, and so when it was sent to fish along the shore, the bird refused 
to work because it had no stamina. 

1. The numberH .show the generations back from the year 1900. 

2. In the times of Vaitakere we learn from other documents that the people 
were living in that Avaiki which has been identified with either Java or Sumatra. 

3. Te Irapanga is in all probability the navigator shown in our ** Memoirs," 
Vol. rV., p. 32, who led the migration from Tawhiti-nui (? Borneo) to the Hawaiian 
Islands, but who apparently settled finally in either the Lau islands of the Fiji 
Group, or in Savai'i of the Samoan Group. 



Digitized by 



Google 



136 JOURNAL OF THE POLYNESIAN SOCIETY, 

255. Theu Tane-au-vaka became angry with the bird, and killed 
it. When Tu-tarangi learnt that both Ids birds were dead, great was 
his anger, and he despatclied his son Etoi to fell the tree named " To 
li-matoa-i-avaiki *' [to make arms of] . So the son went, and felled 
the tree, and when it was down h^ retiirned and reported to liis 
father. The father then said to his son, ** Go thou, and lay the 
matter before [the god, or perhaps a learned man] Tane." 

256. Then Etoi took the wood to Tane, and on his arrival, Tane 
said, ** Return! and tell Tu-tarangi to send hither a priest." In 
accordance with this command, Tu-tarangi sent Rauru-mjioa with an 
offering of food; it was cold (uncooked) food, and was (?) named 
after that tree " Au-makariri." On Kauru-maoa's arrival, Tane 
directed him, saying, " O the powers of earth ! turn over this wood. 
the powers of the land ! split the wood, named ** Te li-matoa-i- 
Avaiki," hew it in pieces, shake it; gnash the teeth, be nimble, glare 
the eyes; that it may return to the breast of Uongo-ma-Tane ! "* 

257. After the wood had been split up, eight weapons were 
dubbed out of it, and the following names given to them : the spear 
was named by Tu-tavake after his own teeth — ** Nionio-roroa " ; the 
aro was named ** Te Aroaro-rangi," the kounga was * Te Pivai-rangi/ 
the mata-tupa was * Te Mata-tua-rero,' the rupo was * Te Poopoo- 
rangi,' the karare (the javelin) * Te Iti-rarerare,' the akatart'-kun (a 
barbed spear) ' Puapua-ai-nano,' the tao (lance) * Rau-tiare.* 

258. After all the weapons were completed they were deposited in 
the house called Oro-kete, which was the ediface at the back part of 
house [? of Tu-tarangi], where was the ataata-itu [? altar] to the god 
Rongo-ma-Tane. When all the weapons had been placed in due 
order, the priest, Rauru-maoa, reported to Tu-tarangi that all was 
complete. Tu-tarangi asked, "Are they really good weapons?** to 
which the priest replied, ** One only is deficient, * Nionio-roroa,' which 
Tu-tavake has placed on the altar of Rongo-ma-Tane.*' 

259. When Tu-tarangi learnt this, he sent a messenger to Kuru 
— his leading warrior — instructing him as follows : " Q-o thou, and 
take possession of the weapons now with Rongo-ma-Tane." In 
consequence Kuru proceeded to the presence of the god, when Tane 
said to him, '*OKuru! Welcome! Kuru, what have you come 
for? You have a strange appearance, thy eyes are (?) staring I '^ 
Kuru replied, ** I have come to fetch the weapons." Tane spoke to 
him, "Enter then!" After Kuru had entered the house all the 
weapons of Tane moved or wriggled ; then he proceeded to examine 
them all, to find which best suited his purpose, even that on the altar 

4. These cryptic sayings are difficult to understand (as were also the Greek 
oracles), but the object appears to have been to facilitate the conversion of the wood 
into Hpears, and to give the latter mana. 



Digitized by 



Google 



History and Traditions of Rarotonga. 137 

of Tane named **Nionio-roroa.'* He decided on that particular one, 
when Tane said, **0 Kiiru! that is a cursed weapon (of evil omen). 
It will be tlie deatli of (the people of) the land, and also destroy the 
land." Kuru replied, " This is the one I choose " (prefer). 

260. Kuru then came forth from the house, and holding up the 
weapon he cast it into the hands of the several gods (i.e., he called on 
those gods to give his weapon power) saying, ** The weapon that shall 
stand in the battle, whose shall it be ? By Kongo, Tane, Eua-nuku, 
Tu, and Tangaroa ! " (shall my weapon be guided). He then seized 
the weapon in his hand, and with it cut off the heads of the children 
of Tu-tavake named Ti-tape-uta and Ti-tape-tai. Tlie boast (accom- 
panying tlie action) resounded afar, when Tane asked, " Kuru ! 
what is tljat ? " ** It is the effect of the sacredness of the weapon." 
^* I said to you, * Kuru ! it is a cursed weapon.* " 

261. Kuru then went out again and met the sisters (of those 
already killed") named Titi-kereti and Tata-kerero, both of whom he 
killed. The sound spread, and then Tane asked, **OKuru! what is 
that?" **That is the woman-consuming power of the weapon!" 
Then said Tane, **It is Tu-tlie-relation-eater. Go, Kuru! I have 
done with it, and do not return ; thine is the tapakau, the rau-ota and 
the moumounga (? wastefulness)." 

262. And then Kuru departed for the other side of Avaiki and ' 
there fought (the people) ; and succeeded in catching Tane-au-vaka 
and all his many men. So Kuru killed Tane-au-vaka. Thence Kuru 
proceeded to andther part and fought there, even unto A mama '• the 
place of Maru-mamao, wlio with his many men fought from daylight 
until tlie evening. (In the battle) Maru-mamao and his party held the 
coast line, while Kuru and his party were by the road side inland, 
and so his eyes were completely })linded (? })y the sun), and tlien 
Maru-mamao struck Kuru in the face with an axe, and killed him. 
Thus the celebrated weapon **Nionio-roroa " became the property of 
Maru-mamao. 

UI-TE-RANGI-OHA AXD DISPERSION OF THE PEOPLE. 

263. There were born unto Etoi (Tu-tarangi's son, the following 
descendants) : — 

57 Etoi 

Etai 
55 Emaunga 

Erangi 

Ui-tamua 

Ui-taringa 
51 Ui-te-raiigir)ra 

5. A mama is one of the islands mentioned in Part V. hereof, as being adjacent 
to Futuna (Home Inland), north of Fiji. But it may be a local name on Savaii, 
where this scene is laid. 



Digitized by 



Google 



138 JOURNAL OF THE POLYMESIAN SOCIETY. 

It was the latter who built a (celebrated) pdi [or sea-going canoe] ^ 
and the timbers of the canoe were men's bones. *^ The keel of the 
canoe (? and the canoe itself) was called ** Te Ivi-o-Atea." The whole 
of the canoe was built of men's bones, and because no bone was long 
enough to form tlie kiato [or connecting supports of the outrigger]^ 
the tree named ** Te Tamoko-o-te-rangi "was felled for that purpose. 
This tree was a reserved (and sacred) tree belonging to Taa-kura and 
Ari. When these two found out that Ui-te-rangiora had cut down 
their tree, they commenced a war with his party and many men were 
killed, but they secured eight portions of the tree, which were dubbed 
into drums, tutunga, [tapa-heating logs] and boards. The drum was 
named **Taka-euua," and was used in the akaariki ceremonies at 
Avarua,^ while the tutunga was named ** Tangi-varovaro." 

264. Then Ui-te-rangiora proceeded to complete his vessel, and 
launched it on the sea. This was the first occasion of seeing the pdi 
and canoes (?of that kind), and the (commencement) of tlie scattering 
of all Avaiki to the various islands. Due to the wars originating in 
Avaiki through Kuru, Taa-kura and Ari, were the people scattered to 
all the islands ; to Avaiki-runga [Eastern Avaiki — Tahiti, Paumotu, 
etc.] to Iti-nui, Iti-rai, Iti-anaunau, Iti-takai-kere [some of the Fiji 
group, no doubt the eastern or Lau islands], Tonga-nui, Tonga-ake^ 
Tonga-piritea, Tonga-mauga, Tonga -rara, Tonga-anue [the Tonga, 
or Friendly Islands] to Avaiki-raro [8avai*i], Kuporu [*Upolu], 
Manuka [Manu*a] , Vavau [North Tonga Group] , Niva-pou [Niua- 
fou] and Niua-taputapu [KeppePs Island, both north of Tonga 
Group] . 

265. Ui-te-rangiora's descendants were : — 

51 Ui-te-rangiora 

Makua -ki- te-rangi 

TeRangi 

Ata-o-te-rangi 

Tara - o - te - rangi 

Te Paku-o-te-rangi 
45 Te Uka-o-te-rangi 

Uu 

Ane 

Taipu 
41 Tuna-ariki 

6. Thus the original reads, but it is a strange statement which has some 
meaning not apparent. Perhaps the canoe was ornamented with bones let into the 
wood. 

7. It would be interesting to learn where this Avarua is, either in Savni'i, 
*Upolu, or perhaps the Lau Group. It is mentioned in several traditions, and is 
not that one at Rarotonga or Ra^atea. The akaariki is the appointment of a high 
chief to his office. 



Digitized by 



Google 



History and Traditions of Rarotonga. 139^ 

In Tuna-ariki's time a war commenced between him and Tu-ei- 
puka, about Avarua. Tuna-ariki insisted that it belonged to him, 
whilst Tu-ei-puka equally claimed it. So Tuna-ariki killed 
Tu-ei-puka, and the au^ [the chieftainsliip] devolved on the former. 
In the end he was killed by a pig, an uru-kivi [? striped] pig, which 
ate that ofiki. 

266. After his death the government devolved on Kati-ongia,^ 
about whom is the saying, **Kati-ongia became ruling cliief, and 
Kuporu (JUpolu) ruled." He was a son of Tu-ei-puka who had a 
brother named Maru. Kati-ongia's * descendants were : — 

40 Ejiti-ongia 

Kapua 

Atonga (also known aa Otenga-atua and Tauira-rangi-o-Avatea). 
37 Te Aru-tanga-uuku 

ABOUT THE VESSEL OF TE AKU-TANGA-NUKU. 

267. Te Aru-tanga-nuku very much desired to have a canoe of 
his own; he was incited thereto by his parents (probably uncles) 
Oro-keu, Oro-i-nano, and Oro-taere. The reason of tliis strong 
desire was the scarcity of food, for the food allowed them by Atonga 
[the ruling chief who is said to have had two natures, one a spirit 
(vaerua) and the other a physical one {kopapa-tangata)] was very 
deficient, very little was given to them or their child (? nephew). 
Hence they incited the elder son (of Atonga) — Te Ara-tanga-nuku to- 
build a vessel in order that they might go to other lands [than 
Kuporu, or 'Upolu] . 

268. After their minds had been made up, they prepared the 
axes, made the customary feast, and next morning shouldered their 
axes and proceeded to prepare a tree as a keel for the vessel. They 
went to the mountains, where they met a ruru [the white heron] and 
an aa [snake]^ striving together. The runt said to Oro-keu, **0 tlie 
chief! Oro-keu! Separate (or end) the fight of the ruru and the^ 
««." The aa said, ** Tlie scarlet-belted cliief must go on his way and 
leave the ruru and the aa to their mutual struggle." And so Oro-keu 
went on his way. 

269. After the above appeared Oro-taere, to whom the ruru 
addressed himself, "0 the chief! Oro-taere! end the fighting 
between the ruru and the aa ! Now Oro-taere felt sorry for the ruru 

8. This Kati-ongia is no doubt identical with *Ati-ong^e of the Samoan 
genealogies. 

9. Snakes of a harmless kind, but sometimes over 12 feet long are found in 
Samoa ; but this was the sea-snake, or pm. For the cause of this strife,* see 
** Rarotonga Records,** p. 83. 



Digitized by 



Google 



140 JOURNAL OF THE POLYNESIAN. SOCIETY. 

because lie was an elder lnother of his, one of Ore's children.^ ^ His 
anger grew towards the aa^ for he supported the cause of the rwrw, 
and lie therefore cut down the aa with his axe, and then lifted down 
his relative the ruru, and, placing it in front of him, wept over him, 
and healed the wounds made by the aa As soon as this was accom- 
plished, the ruru asked Oro-taere, ** What is your object here ? " "I 
aui going to fell a tree to make a vessel for the arikt^ Then said the 
ruru, ** Go to my tree at Ara-punga-verevere ; I did not tell Oro-keu 
and Oro-i-nano about this [because they would not help me in my 
struggle with the <?«]. Probably they are dead on the ridge by this 
time." So Oro-taere went on and found the tree, a Maota-mea was 
the kind, ^ ^ which he felled and commenced shaping out [as a canoe], 
then fixed the hauling ropes {kaka) and left it. 

270. Now, at this time there came Tangaroa-iu-mata [? the 
owner of the forest] and behold ! there laid the fallen tree. He asked 

[to himself], '*Who has fallen my tree?" But he could not find 
out ; so he went to the guardian of the place — Rata-i-te- vao ^ ^ — and 
asked him, ** Who has been felling one of my trees?" Kata replied, 
"I do not know!" Tangaroa then proceeded to enquire of every 
one who dwelt near those parts, Titiri, Tata-rara, and Tu-enua-i-te- 
vao-tere, but they all replied they had no knowledge of the 
•circumstances ; and then he came to the conclusion he would not be 
a})le to ascertain who was the delinquent. So Tangaroa returned to 
the fallen tree, and re-erected it, saying to the tree, ** Stand up thou 
maota-mea, be erect, be girded on thy bark." At this the bole of the 
tree stood erect again, and then he addressed the top [tamoko) of the 
tree, saying, ** Stand there, O thou head of the tree! the large and 
small branches of the tree ! the chips and the leaves return to your 
places ! Adhere, gird on, the bark! " At this the tree stood erect as 
it was previously, and Tangaroa-iu-mata returned to his home. 

271. Sometime after the above, Oro-taere and his partj' returned 
to their work, and on arrival at the the place where the stump ought 
to have been — it was not there ; the tree stood erect ; nothing but the 
hauling ropes suspended on a tree were to be seen. They searched 
and then found the tree by a white place from which a piece of bark 
had been taken down to the sea by them [when they felled the tree] 

10. This looks like the Samoan custom of claiming relationship with certain 
birds, etc. — a species of incipient totemism But probably it was a fight between 
two clans, of whom the heron and sea-snake were the totems, or gods. 

11. Maota-mea is the Dyaoxylon alUacmmy a veiy handsome tree that gi-ows in 
the Samoan Group. There was a beautiful clump of these trees in front of Robert 
Louis Stevenson's house at Vailima, on the hills behind the town of Apia, the 
lower branches of which he had cleared away, learing a charming view of the 
town and the sea under the upper branches, as seen from the house. 

12. Rata-i-te-wao is known to Maori traditions. 



Digitized by 



Google 



History and Traditions of Rarotonga. 141 

ill consequence of tlie sacredness of the tree [and over which cere- 
monies to remove the tapu of the tree had to he performed]. They all 
returned to the shore, and again Oro-taere consecrated his axes, the 
tapu of which had heen destroyed in killing the aa, and hence it 
became j)08sible for Tangaroa to re-erect the tree again. After this 
had heen done they returned and again felled the tree, barked it, 
fixed the hauling ropes, and commenced dragging the log to the 
place where Atonga the priest lived. 

272. The ^^ food was prepared for the priest, for Atonga, the^ 
riaria and the other parts were cooked, but the heap of wood was left, 
it liad not been sliaped. On another day Te Ara-tanga-nuku said to 
his wife, Pori-o-kare, " You must go and take some papaia, [pounded 
and baked taro'] for the priest." She proceeded to cook some, and 
before long took it to Atonga, and after he had eaten and was satisfied,, 
he said to Pori-o-kare, ** Return, and say unto the artki he must 
build a house. To-mon*ow the vessel will be completed, and when 
the house is finished let all Kuporu be seated there so they may 
behold tlie vessel being dragged along by the birds." 

273. After Pori-o-kare had departed, Atonga summoned Tupua- 
ki-Amou ^^ and said to him, ** Haste, and say to the furUy it is to go ta 
the Pirake-akaruirui-rangi, ^ * and assemble all the birds to come and 
drag the vessel of the artki. ^^ He went off and gathered all the 
many birds. When daylight came all the birds surrounded the vessel,, 
the moamoa [? ground birds] took hold of one side, others helped from 
inside. The Kakaia, the Ngoiro and Katikatika families, the quick 
flighted birds on the out side. And tliey said, **With the wings 
strike the stern ; shake the log; lift it; shake the bows; together, 
hasten the ' Ivi-o-Atea.' Gathered together are the many of Kuporu, 
to see the sight; thou will win O Oro-keu ! Oro-i-nano." It was 
tlie Kati-rori bird who recited the song. And now the canoe arrived 
at the house built by tlie hands of Te Ara-tanga-nuku. Tupua-ki- 
amoa had been sent to fetch the vessel, but he failed througli want of 
food. [Another story says that he took off the figure-liead of the 
canoe and Iiid it, but Atonga recovered it.] 

274. This is the explanation about that vessel : It was dubbed 
out in the night by Atonga-vaerua [ Atonga- the-spirit] and his work- 
men, and these are the names of those shipwrights : lu-mata, Aa-ngu, 

13. There seems as if a part of the story had been ommitted here. Riaria 
means refu.se, but the sentence is apparently incomplete. The riaria^ we learn 
from another narrative, were demanded by Tupua-i-Amoa, as his perquisites, but 
the woman refused to give them, and hence Tupua's subsequent action in taking- 
off the figure-head of the canoe. 

14. Tupua is an old family name in Samoa. Amoa is a place on the north- 
east coast of Savai'i Island. 

15. Pirake is a bird noted for its soaring habits. 



Digitized by 



Google 



142 JOURNAL OF THE POLYMESIAM SOCIETY, 

Na-ora, and Na-oti, who built one side of the vessel, while Tupa, 
Tupa-ake, Tupa-aki, and Uri-reka built the other side; there were 
^ight >)uilders, Atonga being the ninth. A tonga named the vessel 
**Taraipo" [built-in-the-night],*^ while the birds also gave it a 
name, **Te Manu-ka-tere." ^ ^ When the canoe readied the home of the 
ariki^ the birds returned inland, but Atonga stopped tire ruru, and 
asked it, *' Where is a tree suitable as a rakau-inkava for the ariki ? " 
The ruru replied, **At Te Po-amio." *' Where is thatx^lace?" said 
Atonga. ** At the Ara-pungaverevere, at the place where I live." And 
then the bird went away. 

276. Atonga* now insti'ucted Tupua-ki-Amoa, saying, ** Go thou 
And cut down Te Po-amio," and explained where it was. So Tupua 
went ofE to enquire of Rata-i-te-vao (the guardian of the forest), 
43aying, " Where is Te Po-amio ? " ** Beyond there," said Eata, so Tupua 
went on and inquired of Tupi-riri, who replied, ** Further on." So he 
went on, but could not find it. Then he descended to Tupa-raro and 
asked him, who said, ** A long way on" ; but he still did not find the 
place. He then went to Tu-enua, in the great forest, who explained to 
him, ** There it is." Then he went on and searched, found it, and 
-cut down the tree which was named Ipi-rere. He brought it down to 
the village and shaped it, and on completion named it "Te Amio-enua." 
He then delivered it to the ariki, who took the weapon and placed it 
in the canoe, and then the vessel was named " Te Pore-o-kare." 

TB AKA-TANGA-NUKU'S VOYAGES. 

276. The vessel was now launched into the sea, and proceeded on 
its voyage to Iva (the Marquesas Islands). The name given to the 
vessel at Iva was **Te Orauroa-ki-Iva " [the long voyage to Iva]. 
From there it went to Eapa-nui [Easter Island] and on to Eapa-iti 
for Oparo, south-east of Earotonga in lat. 27° 30' south, another name 
for which is Eapa-hue], where Irei (or Ivi) was left on account of 
his bad navigation of the vessel. From there they sailed to Avaiki- 
runga [Tahiti and neighbouring groups] and all the islands near 
there. At Avaiki-runga the vessel received a further name, **Te 
Ara-ki-Avaiki." 

277. The great desire of the ariki — Te Ara-tanga-nuku — and all 
the crew, on the completion of the vessel, was to behold all the 
wonderful things on the ocean which had been discovered and 
reported by Ui-te-rangiora [see par. 263] the man's-bones canoe 
(Te Ivi-o-Atea) in former times. The following were those things: 
The rocks growing out of the sea beyond Eapa Island ; the monstrous 

16. Compare the New Zealand Maori story of the canoe of the same name, 
and built under somewhat similar circumstances. 

17. A name by which, I was informed in Tahiti, this celebrated canoe was 
known in that island. 



Digitized by 



Google 



Hhiory and Tradiiion$ of Rarotonga. 143 

waves; the female dwelling in those waves, with her hair waving 
and floating on the surface of the ocean ; and the tai-uka-a-pia [the 
frozen sea], the deceitful animal seen on the sea, which dived helow 
the surface; a very j^looniy and dark place, where the sun is not seen. 
There is also there (a kind of) rock whose summit pierces the sky with 
steep hare cliffs, where vegitation does not grow. Such was the work 
of this vessel at that time; and also to convey people to all the 
islands. It was this vessel, ** Te Ivi-o-Atea,** that discovered all 
these gp*eat and wonderful things on the ocean, and all the suiTOun ding 
islands. 

[The inference to be drawn from the foregoing statement is, that 
Te Ara-tanga-nuku followed in the footsteps of the other great 
navigator, Ui-te-raiigiora, who flourished fourteen generations, or 
350 years before him, and that he visited, some at least, of the 
** wonders " discovered by Ui-te-rangiora, in the seventh century 
(using tlie generation herein given as chronology). There can be 
httle reasonable doubt that the ** wonders'* described above refer to 
the Antarctic regions, ** a very gloomy and dark place where the sun 
is (rarely) seen/' the '* rocks whose summits pierce the sky," being 
icebergs, while the ** deceitful animal " is probably either a walrus or 
sea-lion, while the ** hair waving on the surface " is probably the 
hull-kelp, which these people would not see in the tropics. The 
** Tai-uka-a-pia " is the ice, or frozen sea, like pia, the scraped 
arrowroot, which is exactly like snow, and is just the kind of 
description these people would give to snow or ice, with which they 
would not be acquainted with, except perhaps traditionally. Uka is 
the equivalent of tlie Maori hnka, ice, frost, snow. Such, expressed in 
the poetical language of the islanders of the tropics, is the description 
of the regions south of Rapa, where tlie ice is frequently to be found 
about latitude 50°. The Tongans have also traditions of the frozen 
ocean, which they had visited in ancient times.] 

278. Te Ara-tanga-nuku had the following descendants : — 
36 Te Arn-tanga-rangi 

Te Amaru -ariki 

Te Aniaru-eima 

Te Ueng»-»riki 

Te Uenga-enuA 

Kau-tea 
30 Kau -mango Ono-kura =: Te Ata-ima 

Vai-iti Nga-upoko-tunia 

Knu-kura Nga-maru 

I ^__ Kotnku-tea 

Pou-vananga-roa-ki-Iva =: Rua-mano (f) Kan-ngaki (f) Maonga (f) =: Pou-tea 

, 1 -'■-, I ! 

26 Maono Keu Raka-nui (f) 26 TANGiiA-Nin Tu-tapu 



Digitized by 



Google 



144 JOURNAL OF THE POLYNESIAN SOCIETY, 

[Here we leave these geuerations of adventurous voyagers and 
emigrants, for in the times of Tangiia-nui the twenty-sixth generation 
from the year 1900, we enter upon an important epoc in Polynesian 
history. The time covered by this Part VI., saw the spread of the 
so-called " Tonga-fiti " branches to nearly all parts of the Pacific, and 
it was four generations after Tangiia that the last migration to New 
Zealand took place, i.e., in the middle of the fourteenth century.] 



NO TANGIIA E TE NGA-TAITO-ARIKI. 

253. E tuatua no Te Nga-taito-ariki ; e taraaiti na Te Tumu, anau 
akera tana ko Te Nga-taito-ariki : — 
95 Te Tumu 

Te Nga-taito-ariki 

Mua 

Eanga 

Maina 
90 Maeata 

Makaro 

Te Marama 

Te Eanga -ki-te-ao 

Tangata-kato 

Te Atu-taijganga 

Te Atu-te-ngangata 

Te Atu-te-ki 

Te Atu 

Tiki 
80 Taito-rangi-ngunguru 

Taito-rangi -ioio 

Taito-kuru-angiaugi 

Taito-kuru-ma-raka-mea 

Vai-takere 

Te Tarava 

Te Tarava -enua 

Te Rua-enua 

Te Rua-mata-iko 

Te Punupunu, Ko U'ta-rei(euangaika ia) Auau akera ta Rua-mata-iko 
70 Ara-kapua 

Te Ara-o-iiga-atua 

Toi 

PungH-verevere-ka-ngae-ki-te-ara-o-nga-atua 

Mauu 

Manu-kaiaia 

Mauu-kavakevake 

Ore 

Turanga, ma Te Ruru (e maim ia) 

Rongo-ruH 

Rira 

Te Ira-panga 

Tu-tarangi 
57 Etoi 



Digitized by 



Google 



History and Traditions of Rarotonga. 145 

254. Kua akatupu a Tu-tarangi i te tamaki ; tera te niea i tupu 
«i taua pekapeka ra — e puke inanu na Tii-tarnngi, ko Aroa-uta, e 
Aroa-tai ; e pnke manu rave angaanga nana ; e rave kai nana, 
e tautai ika nana. £ tae ake ra ki tetai tuatau, kua unga 
maira a Tane-au-vaka i te karere i aua nga nianu ra ki a 
Tu-tarangi, kia omai aia i aua nga manu ra nana. E kare akera i 
tika i a Tu-tarangi. E, no te putupntuanga i te tiki mai, kua oatu 
aia i te manu noo uta na Tane-au-vaka — koia a Aroa-uta. Kare i keu 
ki te angaanga, kua ta aia i te reira manu. E kua tiki akaou mai i 
tetai — kua pati rai ki a Tu-tarangi, e kua oatu a Tu-tarangi i a 
Aroa-tai ki aia. E, kua rave taua manu ra i tana angaanga, kua 
tautai i te ika nana. E kare ra aia i takiiiga-ineitaki i taua manu 
ra; kare i angai ki te kai ; e kia tono ra aia i te manu ra kia tiki i te 
ika ki tatai, kare te manu e keu. No te mea, kare ua e kapenga a te 
manu. 

255. Kua riri iora a Tane-au-vaka i te manu, e kua ta iora, mate 
iora. E kia kite ra a Tu-tarangi e, e kua mate nga manu, kua tupu 
iora te riri o Tu-tarangi ; kua tono atura i tona tamaiti, i a Etoi, ei 
tipu i te rakau, i a Te-Ii-matoa-i-Avaiki. Kua aere atura te tamaiti, 
kua tipu atura i te rakau ; e topa iora ki raro, kua oki maira e 
akakite ki te metua. Kua karanga atura te metua ki te tamaiti, 
**Ka aere! kavea ki a Tane." 

256. Kua aere atura a Etoi, kua apai i te rakau ki a Tane. E tae 
atura ki a Tane, kua tuatua maira a Tane, ** E oki ! e karanga atu ki 
a Tu-tarangi, kia unga mai aia i tetai taunga." Kua akaunga atura 
a Tu-tarangi ia Rauru-maoa ; kua keri iora i te kai ei taonga. E 
kai makariri i topaia i te ingoa o taua rakau ra, ko Au-makariri. E 
tae atura a Rauru-maoa, kua tuatua maira a Tane, na-ko-maira, '*E 
te atu papa e ! e kia uriuriia akera te rakau nei, E te atu-enua ! kia 
vavai akera te rakau nei, ko te li-matoa-i-Avaiki ; kia tutuki, kia 
ungaunga, kia ru. kia tete, kia ngavari, kia inana. Kia koki ra e, ki 
roto i te unia (?rima) o Ronga ma Tane e." 

257. E oti akera te rakau i te vavaiia, kua tarai iora e, e varu 
rakau. Teia te ingoa o taua au rakau ra ; Ko Te Tokotoko ; koia tei 
topaia e Tu-tavake ki te nio nona — koia a Nionio-roroa. Ko Te 
Aroaro-rangi ; koia te aro. Ko Te Pivai-rangi — koia te kounga. Ko 
Te Mata-tua-rere — koia te mata-tupa. Ko Te Poopoo-rangi — koia te 
rupo. Ko te Iti-rarerare — koia te korare, e Bau-tiare — koia te tao. 
Ko Puapua-ai-nano — koia te akatara-kuri. 

258. Kua ma te au rakau kua kave ki roto i te are, ko Oro-kete — 
koia te orau i te tuaroa o te are; ko te ataataitu la o Kongo ma Tane. 
Kia oti te au rakau i te akapapaia kua aere atura te taunga ra, ko 
Rauini-maoa ki a Tu-tarangi, **E rakau meimeitaki ainei ? " Kua 
karanga atura aia, ** Okotai rai taka i te rakau ; ko Nionio-roroa tei i 
a Tu-tavake, tei runga i te ataata-itu o Rongo ma Tane." 



Digitized by 



Google 



146 JOURNAL OF THE POLYNESIAN SOCIETY. 

259. E kite akera a Tu-tarangi i tei reira tuatua, kiia tono atura 
aia i te karere ki a Kuru — ko tona ia tumu-toa — kua iia-ko-atura a 
Tu-tarangi ki aia, *' Ka aere koe, ka tiki i te are rakau i o Hon go ma 
Tane." Kua aere atu a Kuru, aere atura, e tae atura ki o Kongo ma 
Taiie. Kua kapiki maira a Tane, ki a Kuru, na-ko-maira, ** E Kuru 
e ! ina ! ka oro mai, E Kuru ! e aa te aerenga. Kua tu ke koe E 
Kuru ! Kua ioi o niata." Kua karanga a Kuru ki a Taiie,» ** E tiki 
au i te are rakau." Kua kapiki maira a Tane ki a Kuru, ** E na roto 
mai.'* E. tei to Kuru toiiioanga atu ki roto, kua keukeu atura te are 
rakau a Tane i a Kuru ; e tae atura aia ki roto kua akara aere iora 
aia i te au rakau ravarai e tau i aia ; mari ra ko te rakau i runga i te 
ataata o Tane, ko ** Nionio-roroa." Kua kiriti maira aia i taua rakau 
ra; kua kapiki maira a Tane, ^' E Kuru e ! e rakau kanga. Ka mate 
te enua, ka nina ua te enua." Kua karanga maira a Kuru, " Ko 
taku rakau rai teia ! " 

260. Kua tomo atura a Kuru ki vao, kua akatu akera i te 
rakau ki runga, kua titiri atura i te rakau ki te rima o te au atua, 
na-ko-ati^ra, ** Ko te rakau e tu i te taua naai ? Na Rongo, na Tane, 
na Rua-nuku, na Tu, na Tangaroa." Kua opu aia i te rakau, kua 
taki akera ki runga i tona rima, kua tipu iora ki nga tamariki a Tu- 
tavake — i a Ti-tape-uta e Ti-tape-tai. Kua vavaro atura te iio ; kua 
ui maira a Tane, ** E Kuru e ! ko te aa tena ? " ** Ko te tapu tena i 
te rakau," " 0, i karanga atu na au ki a koe, * E Kuru e ! e rakau 
kanga ! ' " 

261. Kua oki akaou atura a Kuru, kua aravei atura i nga tuaine, 
i a Titi-kereti, e Tata-kerero ; kua mate ia tokorua ; kua vavaro mai 
te iio ; kua ui atura a Tane, " E Kuru ! ko te aa tena ? " " Ko te kai 
vaine teia i te rakau." Kua na-ko maira a Tane, " Ko Tu-kai-taeake 
tena. E Kuru ! e aere ; kua oti taku. Aua e oki mai. Naau atu te 
tapakau, e te rau-ota, e te moumounga." Kua aere atura a Kuru ki 
tetai pae i a Avaiki, kua tamaki atura ; kua rauka iora taua ariki ra 
ko Tane-au-vaka ma tona tini tangata katoa. E kua ta iora a Kuru i 
taua ariki ra, i a Tane-au-vaka, mate iora. 

262. Kua aere atura aia — a Kuru — ki tetai pae, kua tamaki. E 
tae ua atura ki Amama ki o Maru-mamao. Kua tamaki maira a 
Maru-mamao ma tona au tangata i aia i te akirata i te popongi. Ko 
Maru-mamao ma tona au tangata, i a ratou a tai — i te tapa tai ; ko 
uta a Kuru i te tapa-ara. Kua tamaki iora, kua verovero maira te 
ra ki nga mata o Kuru ; kua poiri kerekere iora nga mata o Kuru. 
Kua pari maira a Maru-mamao i nga mata o Kuru ki te toki ; mate 
iora aia. Kua riro atura taua rakau ra i a Maru-mamao. 



Digitized by 



Google 



History and Traditions of Rarotonga. 147 

263. Anau akera to Etoi, ko : — 

Etai 
55 EmaungH 

Erangi 

Ui-tamua 

Ui-tariuga 
51 Ui-te-rangiora 

Kua tarai aia i te pal, e ivi taiigata te rakau i taua pal ra. Ko te 
ivi o Atea te takere i taua pai ra. E oti ua ake taua pal ra e ivi 
tangata anake. No te mea ra kare e ivi tangata roa ei ova, ei kiato,. 
no reira i kotia ei a te **Tainoko-o-te-rangi" ei kiato, ei ova. E 
rakau raui na Taa-kura e Ari. E kite akera a Taa-kura e Ari e, 
kua motu taua rakau ra, i a Ui-te-rangiora ; kua tamaki atura, kua 
mate iora te tangata, kua riro maira te rakau i a raua. E ono- 
potonga i riro mai, kua tarai iora ei pau, ei tutunga, ei papa. Ko te 
ingoa i te pau ko Taka-euua — koia te pan akaariki ki a Avarua. Ko- 
te tutuiiga ko Tangi-varovaro. 

264- Kua rave akaou a Ui-te-rangiora i te pai, kua akaova, e- 
kua ri, kua aau i te kiato ; kua oti, kua tuku ki te tai. Ko t& 
kiteauga akera rai la i te pal e te vaka. Ko te pueu-rikirikiuga teia 
i a Avaiki ki te pa euua. No te tamaki i tupu i Avaiki e Kuru, e- 
Taa-kixra e Ari, kua pueu-rikiriki atura Avaiki ki te pa-enua 
ravarai ; ki Avaiki runga, ki Iti-uui, ki Iti-rai, ki Iti-anaunau, ki 
Iti-takaikere, ki Tonga -nui, ki Tonga-ake, ki Tonga-piritia, ki Tonga- 
manga y ki Tonga- raro, ki Tonga-anue, ki Avaiki -raro, ki Kuporu, ki 
Manuka, ki Vavau, ki Niva-pou, ki Niua-taputapu. 

265. Anau akera ta Ui-te-rangi-ora ko : — 

50 Makua-ki-te-rangi 

Te Rang! 

Ata-o-te-rangi 

Tara-o-te-rangi 

Te Paku-o-te-raugi 

Te Uka-o-te-rangi 

Uu 

Ane 

Taipu 
41 Tuna-ariki 

Kua tupu te tamaki i a raua ko Tu-ei-puka ; Tera te ara ; ko- 
Avarua. Te manono nei a Tuna-ariki nona a Avarua ; te manono mai 
a Tu-ei-puku nona. Kua ta iora a Tuna-ariki i a Tu-ei-puka ; mate 
iora. Kua riro maira te au ki a Tuna-ariki. I te openga iora kua 
mate aia, kua pou i te puaka, e uru-kivi te puaka i keinga i (? ai) 
taua ariki ra. 



Digitized by 



Google 



148 JOURNAL OF THE POLYNESiMK SOCIETY. 

266. I muri iora i aia, kua riro te au ki a Kati-ongia — i tuatuaia 
ai e, "Kua ariki Kati-ongia, kua au Kuporn." E tamaiti aia iia 
Tu-ei-puka. E teiua a Maru no Tu-ei-puka : — 

40 Ko Kati-ongia, anau tana ko * 

Kapoa 

Atong^ (known also as Otenga-atoa and Tauira-rangi-o-avatea) 
37 Te Ani-tang^-nnkn 

NO TE PAI O TB ARU-TANGA-NUKU. 

267. Kua akakoro a Te Aru-tanga-nukii ei vaka tona. No nga 
metua tane te manako no Oro-keu e Oro-i-nano e Oro-taere. Tera te 
tupuanga i taua manako ra, e aue kai. Ki te kai a Atonga e kai lui 
maira, kare e omai na ratou ma ta ratou tania. No reira ratou i 
akakoko ei, ki te tama ariki — ki a Te Aru-tanga-nuku — kia tai-ai i 
tetai pal ei ara no ratou ki te pa-en ua. 

268. E taka akera te tuatua, kua rango iora i te toki ; kua moe i 
te angai e popongi akera, kua apai te toki ki te tipu i te rakau, i te 
takere i te paT. Kua aere atura ratou e tae atura ki te maunga, kua 
aravei iora ratou i te Ruru e te Aa, te taiapiapi ua ra. Kua kapiki 
maira te Ruru ki a Oro-keu, ** E te ariki, E Oro-keu ! e vaoa te taua 
a te Huru ma te Aa.*' Kua tuatua maira te Aa, '^ E aere rai te ariki 
Maro-kura i tana aere e vao rai te Ruru ma te Aa kia taiapiapi 
marie." Aere atura aia i tana aere. 

269. Kua mama atura a Oro-taere ; kua kapiki maira te Rnru, 
** E te ariki, e Oro-taere ! a vaoa te taua a te Ruru ma te A a." Kua 
tupu akera te aroa ki roto i aia, no te raea, e tuakana te Ruru nona — 
ko tetai tamaiti la a Ore. Kua tupu akera tona riri ki te A a, kua 
tauturu atura aia i tona Uieake, kua tipupu iora i te Aa. E mate 
atura i taua taeake nona ra, ki )*aro, ki mua i aia, kua aue iora ki 
runga i te Ruru. E oti akera kua rapakau iora ki te vai, kia papa te 
etietinga a te Aa i aia. E oti akera tei reira, kua ui maira te Ruru 
ki a Oro-taere, **E aa to aerenga ? '* ** E koti rakau toku aerenga; e pal 
no te ariki.*' Kua na-ko maim te Ruru, ** E oro ki taku rakau, i te 
Ara-pungaverevere ; kare au i akakite ki a Oro-keu e Or*)-i-nano. 
Tera raua kua mate ki runga i te kaivi.** Kua aere atura a Oro-taere 
e tae atura ki taua rakau ra, e Maota-mea te ingoa i te rakau. Kua 
tipu iora, e kua pari, kua tamou te kaka, e vao kia vai. 

270. Kua aere maira a Tangaroa-iu-mata, e ina ! kua motu te 
rakau. Kua ui maira na-ko maira. " Naai i tipu taku rakau ? *' E 
kare akera i kitefi ; kua aere atura ki te tiaki i te kainga — ki a Rata- 
i-te-vao — na-ko atura, "Naai i tipu taku rakau?*' Kua karanga 
maira, ** Kare au i kite." Kua ui tatakitai aere atura ki te au 
tangata ravarai i vai tata mai ki taua kainga ra, ki a Titiri, ki a 
Tatarara, ki a Tu-enua-i-te-vao-tere, kua na-ko-niaira ratou, ** Kare 
rava niatou i kite.*' E kite akera aia e, kare i kitea. Kua oki atura 



Digitized by 



Google 



Hisiory and Traditions of ffarotonga/ 149 

aia — a Taiigaroa-iii-mata — kua akatu akaou i te rakau, e kua kapiki 
atura ki te rakau, na-ko-atura, ** E tu te maota-niea e, ka tu mai ki 
riiiiga ; kia kiri ; kia taka.*' Kua tu maira te tumu ki runga i reira ; 
kua kapiki akaou atura aia, ki te tamoko o te rakau, na-ko atura, 
** Ka tu mai koe, e te kauru o te rakau ! e te manga o te rakau * e te 
atava o te rakau ; e te vara o te rakau, e te uugaunga o te rakau, e te 
ran o te rakau. Kia piri ! kia kiri ! kia taka ! " Kua tu te rakau ki 
ruuga, kua aere atura aia ki tona kainga. 

271. E niiriugao i te reira, kua aere atura a Oro-taere, e tae 
atura aia ki te tumu i te rakau — kare ua ; kua tu ki runga. Ko te 
kaka ua tera e tarava ua ra ; kua kimi iora ratou, e kitea iora ki te 
ngai kiri e tea ua ra ; ko te iigai kiri la i tuoia ki tai, i te tapu 
i te rakau ki o te iiotanga. Kua oki atura ki tai, kua rango i te 
toki, no te mea, ko te ara la i akatuia te rakau no te toki i tipuia ki 
te Aa — kua noa ki reira te toki. E kia oti, kua oki ratou, kua tipu i 
te rakau ; kua motu ki raro, kua pari, kua ma ; kua tamou te kaka ; 
kua kika ; kua to to atura ki o te taunga, ki o Atonga. 

272. Kua tau atura i te kai na te taunga — na Atonga. E ope ua 
atura nga riaria e te enua i te tau atu, te vai ua mai rai te ututua 
rakau, kare akera rai i taraiia. E tae akera ki tetai ra, kua tuatna 
atura a Te Aru-tanga-nuku ki tana vaine, ki a Pori-o-kare, ** Ka aere 
koe, ka rave i tetai papaia na te taunga." Kua aere atura aia, kua 
tau ; e maoa iora, kare i maniia kua apai atura ki a Atonga ; kua kai 
iora aia, e pangia akera, kua tuatua maira aia ki a Pori-o-kare, na-ko 
maira, "*Ka aere, ka akakite atu ki te ariki, kia anga i te are; Apopo 
kua oti te pal. E kia oti te are, akanooia a Kuporu ki raro, kia 
matakitaki i te pal i te totoanga mai a te manu." 

273. Tei te aerenga o Pori-o-kare, kua karanga atura a Atonga 
ki a Tnpua-ki-Amoa na-ko atura, " Ka oro koe, e karanga atu ki a te 
Ruru ; kia aere koe ki a te Pirake akaruirnirangi ei oro i te maiiu 
tini ei kika i te vaka o te ariki." Kua aere atura aia ki te oro i te 
manu, e katoa akera te manu tini ; kua tae ki te popongi, kua iri te 
manu ki runga i te vaka; kua pakipaki iora te aronga moamoa na tai 
pae, na tai pae, i te papaki i te manu ki roto i te vaka. Ko te Kaknia, 
e te Ngoiro, e te Katikatika — ko te aronga oro la i vao. Kua kapiki 
iora, na-ko atura, ** Pakia i miri vaka, ueuea te tumu ; ka maranga- 
ranga, ka ruea (? niea) te in o te vaka ; ka tere, ka niaoi*u (? niaeru) 
ki te Ivi-o-Atea. Ka topokipoki e te tini o Kuporu, ka matakitaki; 
ka re koe e Oro-keu e Oro-i-nano; Oro-keu, Oro-i-nano." Na te 
Katirori i tumu te amn. Kua tae atura te vaka ki roto i te are, ta te 
rima o Te Aru-tanga-nuku i rave. Kua unga iora a Tupua-ki-Amoa 
i te tiki i te vaka, tera te ara, e aue kai. 

274. Teia te tu o tana vaka nei ; i taraia i te po, na Atonga- 
vaerua i tarai ma te aronga rima-rave. Teia taua aronga ra : Ko 



Digitized by 



Google 



150 JOURNAL OF THE POLYNESIAN SOCIETY. 

lu-niata; ko Aa-ngu ; ko Naoti — ko te arotiga taunga ia i te rave i tetai 
paeitevaka. KoTupa; koTiipa-ake; koTiipa-aki; ko Urireka — ko 
ratou te rave i tetai pae ; tokovaru taunga, ko Atoiiga ka tokoiva. Kua 
topa iora a Atoiiga i te iiigoa o taua vaka ra, ko Taraipo ; kua tapa 
te luanu i ta ratou iiigoa, ko Te Maiiu-ka-rere. E riro atura te vaka 
ki o te ariki kua oki te maim, kua tapu atura a Atouga i te Ruru, kua 
ui atura ki aia, ua-ko atura, ^^ Tei ea ake te rakau ei rakau tukava na 
te ariki?" Kua tuatua maira te Ruru, "Tei te Po-amio." Kua 
na-ko luaira a Atouga, " Tei ea ia iigai ? " " Tei te Ara-puugavere- 
vere, i te ugai taku e noo iiei." E kia riro atura te manu kia aere. 

275. Kua karauga atura a Atonga, ua-ko atura, ki a Tupua-ki- 
Auioa, ** Ka aere koe, ka tipuia te Po-amio." Kua ui akaou atura a 
Tupua-ki-Amoa, "Tei ea ia ngai?" Kua tuatua maira a Atonga, 
"Tei te Ara-punga-verevere." Kua aere atura aia ki a Rata- 
i-te-vao, kua ui atura ki aia, " Tei ea a te Po-a-mio? " Kua ua-ko 
maira aia, "Tei ko atu." Kua aere atu aia ki a Tupi-riri, kua 
karanga maira aia, "Tera atu." Kua aere atura aia ki reira, kare 
rai. Kua topa atura aia, ki o Tupa-rara kua akakite maira aia ki 
aia, " Tera roa ai." E tae atura aia ki reira, kare ua rai. Kua aere 
atura aia ki o Tu-enua, i te vao-tere, kua akakite maira a Tu-euua i 
te vao-tere ki aia, " E tera." Kua aere atura aia, kua kimi ; e kitea 
iora, kua tipu atura. E ko Ipirere te ingoa i taua rakau ra. Kua 
apai maira ki te kaiuga, kua tarai iora, e oti akera kua topa iora i te 
ingoa, ko Te Amio-enua. Kua apai atura, kua tuku ki te lima o te 
ariki. Kua rave te ariki i taua rakau ra, kua aao ki roto»i te pal; 
kua topa iora i te ingoa o te pai ko Pori-o-kare. 

276. Kua tuku atura i te pa! ki te tai akatere atura ki Iva. Ko 
te ingoa o taua pal ra ki Iva, ko Te Oraur(»a-ki-Iva. Aere atura ki 
Rapa-uui, e Rapa-iti, akaruke iora a Irei (?Ivi) ki reira. No te 
akatere kino i te pai i akarukeia. Me reira, kua aere ki Avaiki- 
runga, e ki te pa enua katoatoa e pini ua ake. Ko te ingoa ki 
Avaiki-runga i taua pai nei, ko Te Ara-ki-avaiki la. 

277. Tera te akakoroanga a te ariki — a Te Aru-tanga-nuku — ma 
te au tangata, i te otinga i te pai, ko te au mea i te moana ko tei 
kitea e te pai ivi-tangata, i muataugana. Tera taua au mea ra ;; 
ko te mato tupu i te moana — tei tai-rua-koko ra, te peru ua ra i te 
rauru i roto i te moana, e i runga i te kiri-a-tai ; e te tai-uka-a-pia ; e 
te puaka pikikaa i runga i taua tai ra, ko tei ruku ki raro i te tai — e 
ngai ave ua e te popoiri, kare e kitea e te ra. Tera tetai, e mato rai, 
kua tae roa te take ki roto i te rangi, e pare-moka ua, kare e 
ngangaere e tupu. Ko te angaanga la i raveia e teianei pal i taua 
tuatau ra, ko te tari i te tangata ki te pa-eiiua ravarai. Ko te 
Ivi-o-Atea te pal i kitea ai te au mea katakata nunui ki te moana, ma 
te pa-enua e pini-ua-ake. 



Digitized by 



Google 



History and Traditions of Rarotonga. 151 

278. Kua anau akera ta Te Ara-tanga-a-nuku ko : — 

36 Te Ara-tanga-rangi 

Te Amam-ariki 

Te Amaru -enua 

Te UengH-ariki 

Te Uenga-enan 

Kau-tea 
30 Kau-maugo ODO-kura ^ Te Ata-nua 

Vai-iti Nga-upoko-^rua 

Kau-kura Nga-maru 

I Kotuku-tea 

Pou-vananga-roa-ki-Iva = Rua-niano (f) Kau-ngaki (f) MaoDga(f):=Pou-tea 



26 Maono Keu Raka-nui (f) 26 TANonA-NXJi Tu-tapu 



Digitized by 



Google 



TRADITIONS AND LEGENDS. 



eOLLECTED FROM THE NATIVES OF MURIHIKU. 

(Southland, New Zealand.) 



By H. Beattie. 



PART X. 
Continued from Volume XX VIII., page 61. 

WHEN the collector of these tra<1itioi)s had gained the confidence 
of the aged soiithem Maoris, information gathered in 
surpiising volume, so there is a good deal of additional matter on liand 
relative to items already published, and as it is mostly from hitherto 
untapped southern sources no apology is required for its appearance, 
although it is somewhat belated no doubt. 

In regard to Te Ilakitauneke* (who flourished about 1660) lie was 
caught by a Kai-Tahu waka-ariki (my informant said this was a 
battalion, while a tana was 'an army) and supposedly killed, and they 
left him on the ground and went on to his /?a, where they found him 
standing in the gateway and they were beaten. That was the work of 
his atna or god named Matamata. When Rakitauneke died (continued 
my informant, who had received his information from Tare Wetere te 
Kahu, a descendant of Rakitauneke) his body was buried, in accor- 
dance wit!) his dying instructions, in a cleft in the rock on the summit 
of the Bluff Hill, with his face to the rising sun, so tliat he could 
overlook Murihiku. Hence the name of that hill is Motu-poua 
(motn = island, poua = an old man). When the narrator was a boy 
there were bones in a crack in the rocks, but lie did not tliink they 
were Rakitauneke's ; but it was a fapu spot until the Pnkehas levelled 
the top of the hill to build the pilot observatory. The famous Tu-te- 
makohu was a descendant of Te Rakitauneke. 

In the account of Te Rakitauneke, as published (Vol. XXIV., p. 138), 
the collector would add that the then naiTator used no fewer than six 
terms in describing Matamata, or the god, four being Maori and two 
being English. The text does not make this quite clear. 

•See ** Journal Polynesian Society," Vol. XXIV., p. 138. 



Digitized by 



Google 



Traditions and Legends. 15$ 

THE FIGHT AT MOKAMOKA.* 

Wlien tlie Kai-Talui chief Waitai, who settled at Mokamoka) near 
the Bhiff, ill the seventeenth century, was killed by Iiis Kaii-Mamoe 
foes, all his men except four shared his fate. Kaiapu and Tamakino 
escaped and' eventually reached Kaikoura, and the other two, Rere- 
whakaupoko and Potowa were not actually in the fight. They were 
coming from the Mata-au (Molyneux river) and saw the defeat of 
their companions and bolted into the seaward bush. They crossed to- 
Buapuke with their wives, and were the first inhabitants of that island. 
Subsequently, twoKai-Tahu men visited them, but insulted their wives 
so the two visitors were slain, cooked, and potted with titi (mutton- 
birds). A layer of tifi was placed in a rimu (kelp bag) and then a 
layer of human flesh and so on alternately. Some of the rest of the 
tribe came to reside on the island, and were regaled with the "potted 
meat." They praised it, asking was it pakake (seal), but when they 
found what it was they threw it away. Altlnmgh some were relatives- 
of the slain men nothing was said, and my informant never heard of 
an}' warfare over the affair. 

AN OLD-TIME DUEL.f 

In regard to the killing of Tu-takahi-kura by Tu-te-makohu, an old 
Maori tells me it did not happen at Taukohu (Nuggets Point), but at 
Paekohu, which is a hill between the Taieri Plain and Blueskin Bay. 
The hill is noted as a weather-glass; fog on it being a sign of rain» 
Another old man said: — ** There were two chiefs called Tu-te-makohu, 
so one is known as Tu-te-makohn-a-Karapohatu, and the other a& 
Tu-te-makohu-a-Korekore. It was the former who pursued Tu- 
takahi-kura when this chief ran off witli his two wives and his 
children. He overtook them at Waitete (now called Waitati), where 
they were encamped, and he camped near them. He did not sleep at 
nights, but sat brooding over his troubles. Then he challenged 
Tu-takahi-kura to a combat, arranging that the latter's men should 
let him go with his wives and children if he killed their chief. The 
latter was a big powerful man, while the challenger was low in 
stature. They fought with the taiaha (also called maipi)^ and 
Tu-te-makohu killed his tall opponent and took his own wives and 
children back to 0-taupiii (Hokanui Hills), while the people of the 
slain chief went on to their home at Kaiapohia." My informant 
added, *' When the Kati-Mamoe were plentiful they were a quiet and 
peaceful people, and the Kai-Tahu did as they liked with them ; but 
once they were reduced in numbers and in land they roused themselves- 
and fought like tigers." 

•See "Journal Polynesian Society," Vol. XXIV., p. 139. 
t See ** Journal Polynesian Society," Vol. XXV., p. 15. 



Digitized by 



Google 



154 JOURNAL OF THE POLYNESIAN SOCIETY. 

RELATIONSHIP OF TE WBRA TO TAOKA* 

In the narrative the relationship of Te Wera and Taoka is given as 
undecided, but one of iny informants quickly dispelled the mist of 
doubt. 

Te Elaue = Haki-te-ao Te Kaue = Haki-te-ao 

! I 

Tauira-ki-waho (m) = Te Ao-taurewa (f) = Ruahikihiki 

I ! 

I I 

Te Wera Taoka 

Listen to my informant : — " Te Kaue was a famous Kai-Tahu chief 
up Kaikoura way. [See Stack's " South Island Maoris," pages 45, 56 
.and 66.] He married Haki-te-ao and had three children, the first a 
girl Te Ao-taurewa, the next a girl Te Hikaiti, and the last a boy 
Tauira-ki-waho, who married a woman whose name escapes me, and 
whose son was Te Wera. Te Hikaiti married Ruahikihiki (a son of 
Manawa's) and begat Te Matauira, Moki and Ritoka. Her elder 
sister Te Ao-taurewa married a man called Te Ao-taumarewa and 
begot Manaia. Then her husband died and she married Ruahikihiki 
and begat Taoka, and her sister became jealous and with her own 
newly-born babe jumped over the cliff at Hakaroa and was killed. 
Therefore Te Wera and Taoka are first cousins because they were the 
children of a brother and sister.*' 

MOKE ABOUT TE WERA.f 

In regard to the warfare between these two famous cousins one of 
my informants made a correction regarding the killing of Taoka's son, 
Roko-marae-roa, by Te Weni. He said : — " It was not Te Wera 
himself who killed the eldest son of Taoka, but his people, and he got 
back in time to assist in eating the body. The place was not at Timaru, 
but near the mouth of the Waitaki river. You can see the spot from 
the railway line, and it is still called Ka-unm-o-Roko-marae-roa." One 
of my informants considered that Te Wera's friend who was killed at 
Mapou-tahi was named Puke-hau-kea. 

Two old men at Stewart Island said, **When Te Wera came to 
Rakiura (Stewart Island) he found the Kaiarohaki pa on the Moutere 
(island) of Turi-o-Whako (near The Old Neck) deserted, as there was 
no water on the island. A pa o\t the mainland near there was Taunoa, 
but there was no one in it. Then Te Wera went round to Pu-tatara 
pa at Raggedy, but there was no one there to figlit. [The inhabitants 
of this pa under Tukete were killed by Tu-wiri-roa two generations 
before this, as narrated earlier in these jirticles] Te Wera died there — 

*See *' Journal Polynesian Society," Vol. XXV., p. 15. 
t See •* Journal Polynesian Society," Vol. XXV., p. 56. 



Digitized by 



Google 



Traditions and Legends. 155 

not; at Kawhakaputaputa, near Colac Bay, as is sometimes said — and 
was buried at Piitatara, a valuable piece of gi*eenstone being buried 
with him. Except Putatara pa being captured by Tuwiriroa, there was 
^lo fighting on Rakiura, nor was there ever any on Kuapuke as far as 
we know." 

WARFARE AT KINGSTON. 

One of Tu-te-niakohu's wives had two relatives named Te 
Papatu and Korapa, and these men were in a party which attacked 
Tn-te-niakohu on the Waimea Plains. The fight stopped when 
Tu-te-Makohu got a spear thrust (by a man named Tawhana) through 
his arm, and Te Papatu and Korapa tended to his wound. Tu-te- 
makohu told them there had been enough fighting, but they persisted 
in going to attack Marakai. They said they would kill the chief and 
fill a kelp-bag with his cooked body. Marakai was living at the 
south end of Lake Wakatipu, so Tu-te-Makohu sent a lad to warn him, 
and Marakai laid a trap for his attackers. He built a takitaki ( a yard 
and fence) outside his house, and laid in a stock of firewood. He told 
his men te tell the visitors he was at the Mataura (where Garston now 
is), but would soon be back. When he saw them coming he went 
inside, and they came to the takitaki and hung their weapons on the 
fencfe. After dark Marakai crawled out of the window of the tvhare to 
where some flax was heaped. This he wound round him, and then he 
sat with his enemies and talked to them. He put wood on the fire and 
then retired while they went to sleep. Next Marakai silently took the 
flax that enwrapped him and tied all the weapons securely to the 
takitaki. Then the killing coujnienced, the suddenly-roused men 
tugging at their weapons and falling easy victims to their wily 
ndversaiy. This killing led to furtlier figliting of which the collector 
has not got the details. 

THE KILLING OF MARAKAI.* 

One of my informants says that the words said by the Marakai to 
Matauira were as follows : — ** Me he mea naku, na ka to ake kanaka ki 
Waipahi ko tenei kua knkure noa atUj^^ and says that the place referred 
to as the "ford of the Waipahi" is on the old road between Clinton 
and Mataura, and even to this day the natives jcall the spot where the 
road crosses the stream Te-kauaka-o-Waipalii. Another of the old 
men tells me that he thinks Marakai and Tu-te-Makohu were related, 
and that Tu-wiri-roa was also connected by Wood ties with both these 
celebrated chiefs. Another of my informants wrote down the last 
remarks of Marakai as follows: — ^'JSe hara i an na te Marama inini ki ka 
ivhetn mea nahaku ka kanaka ki Waipahi kua kukura noa atu. The latter 

♦See '* Journal Polynesiau Society,'* Vol. XXV., p. 54. 



Digitized by 



Google 



156 JOURNAL OF THE POLYKESIMN SOCIETY. 

halt* of these words mean — if it was me at the crossing at Waipahi 
they would be exterminated long ago." Seeing that the collector's 
ignorance of the Maori tongue led him slightly astray on his former 
account of the speech of Marakai, the above represents his mentor's 
efforts to put him right. 

THE CAPTURE OF TAREWAI * 

The narrator said : — **When the Kati-Mamoe caught Tarewai on 
the Otago peninsnla they killed a Kai-Tahu chief and tohuha called 
Kahutupunei. The chief of the Kati-Mamoe was Tiroko-takanewha. 
Stack calls him AVhaka-taka-newha, but the other name is right. The 
practice of ohu is to get 2)eople to help you with work and yoir feed 
them. The Kati-Mamoe got the Kai-Tahu to help build a house. 
After the feed they began to wrestle in fun at a place called Ka-pnke- 
turoto. The atua of Kaliutupunei told him to say, * Told ivhakaruru 
te maie^ and someone said, *it must be that toki that Tiroko has,' but 
Tiroko said it was an axe to chop firewood. The men were wrestling 
in detached lots, hidden from one another, and the Kati-Mamoe killed 
several, and one escaped and gave the alarm. He called out to 
Kahutupunei, * E Kahu J te tohana ' (start a charge), but Tiroko, who 
was sitting beside that tohuha, said, * E Kahu ! tiki ko to tire ' (a vile 
taunt), and struck and killed him. All the rest were killed save 
Tarewai, whom they laid out to cut open with a ^ mata^ or ^ parahi 
pohatu ' (rude stone knife). They had not made much progress when 
his guards, deceived by his quietness, relaxed their hold, and he gave a 
yell and jump and darted into the bush. In the bush he did * tahu- 
tahu^ or put hot fat in his wounds. [See Stack for narrative, page 85.] 
. . . ^' When he made his famous leap he threw his patu up, and it had 
a string on it and this string curled round a kokomuka shrub (called 
also koromiko in North Island), and he clambered up and got into the 
Kai-tahu j[?«. I do not know whether Tiroko was killed or died 
naturally, but Te-waha-o-te-marama an iramutu (nephew) of Tarewai 
was killed with him at Preservation Inlet. The party went round in 
two big double canoes, and Terewai's got there first and anchored off 
tlie Kati-Mamoe pa. A man swam out and tied a rope to the canoe^ 
and it was hauled in and the crew were invited into the pa and 
installed in a whare. The people made an oven and called to Tarewai 
to come out. His nephew was going, but Tarewai stopped him and 
went and was killed after a* brave tight, and the rest were killed easily. 
Tlie other canoe came in next day, and one of the crew acted as a seal 
and they captured the pa and stayed round there." 



*See ** Journal Polynesian Society," XXV., p. 59. 



Digitized by 



Google 



Traditions and Legends, 157 

MORE ABOUT THE DOUBLE CANOES.* ^ 

To the notes supplied by the collector on this subject an editorial 
footnote was appended. The collector must plead guilty to a looseness 
of phraseology, which rightly drew the editorial correction. He was 
aware that some of the canoes of the 1350 migration were double 
canoes, but that the North Islanders allowed this method of seafaring 
to fall into desuetude, whereas the Southerners adhered to it. Since 
his notes were published three old men descnbed these canoes to the 
collector, and he will here give further details. One said, ** In double 
canoes the larger one was called nntia^ and the smaller tawai. The 
space under the deck between the two canoes was called aroa, and men 
got under there and heaved with their shoulders to help the canoe 
being hauled up on shore. The man who swam out to Tarewai's 
canoe made fast the i-ope to the aroa^ Another informant said : — 
" My son was a sailor and has told me of the canoes in the South Sea 
Islands having outriggers, but with our double canoes both were 
proper canoes. Beams were laid across the canoes, and on these 
decking was built between the canoes. The mast did not rifee.from 
either canoe, but from the centre of the decking. The mast was 
called hua^ and now shipmasts are called by this name. Cordage was 
called taura, and sails ra, 1 think this name ra came about because 
when you hoisted the sail against the sun yoii could see the rays 
showing through the ttaka of which it was made. The hollow of the 
canoes was called te-rin^o-te-wakatere^ the paddles were cnlled hoe-tta, 
and tlie steering oars were hoe-whakatere, A steerer usually stood in 
the rear of each of the canoes, and was called a takato^tohakaiere. 
Another old man said :^** The bigger canoe was nnna, and the smaller 
waka ; the platform was orau tvawa, the mast Atm, the sail ra, and the 
seats in the canoes iraku. The beams were put in rua (holes) in the 
sides of the canoes and lashed with whitan (flax). The splice 
underneath the platform was called arowa, [Note — He wrote the 
word down aro-ioa,^ I was once tm a double canoe. The Mata-au 
(Molyneux) was very rough and we wanted to cross. My father and 
brother were drowned in that river, so Kakitapu chopped manuka for 
beams, and made a platform between the two canoes to avoid a capsize. 
It was a rough-and-ready toaku'imna, but it kept us quite safe." 

THE FATE OF TAHUNUNU. 

One of my informants said: — **Tahununu was killed at Waikakalii 
(Little liiver) by a taua from the south. Hinehaka a prophetess on 
Ruapuke Island foretold his doom in these words — * Tahununu i 
Hakaroa e taki ra ki Wuikoan, apopo ia o iwi taki ana ka turaka Potaetu 
ki te pa tele a te hoa o Whakatepe kolahi te ika i kai mat ko taku mako e.'' " 

♦See ** Journal Polynesian Society,*' p. 60. 



Digitized by 



Google 



158 JOURNAL OF THE POLYNESIAN SOCIETY. 

[Which was translated to me as **Tahuiiiinu at Hakaroa you had 
better cry for Waikoau, as by-aud-by your bones will cry on the 
groper-grounds on the fish-hooks of the husband of Whakatepe. One 
fish took my bait — my shark."] " And," continued the old man, **8ure 
enough his bones were brought south and made into liooks and used 
for fishing at Potaetu, the groper-ground off the New River Heads." 
It may be added that Waikoau was Tahunuuu's home, and that 
the husband of Whakatepe was callee Kuao. The affair took place at 
the time of the Kaihuka feud, and Whakatepe died an old woman at 
the Bluff about 1861. 

ANOTHER FAMOUS DUEL. 

Puneke was the youngest son of Tnrakautahi ))y the latter's 
sec<md wife Tawharepapa, Two men from Mokamoka rei)orted that 
the Kati-Mamoe had killed Kai-Taim people there, so a Kai-Tahu 
party went up to Lake Wanaka to kill some Kati-Mamoe in revenge. 
They captured a Kati-Mamoe chief called Eaki-amoauiohia and asked 
who would fight him in single combat. Puneke volunteered, and 
a duel with patu-paraoa resulted. Puneke was a youth and short in 
stature, but wiry and strong, while his opponent was big and heavy. 
They fought on and on with no advantage, until both were so tired 
that they had to have a spell. The sun was lugh when the duel 
started, but it was going down when the spell occurred. After the 
combatants had rested a while the old people said, *^ You had better 
start again as the sun's legs are hanging down" (i.e., over the hole 
where it disappears every night). After this advise Baki advanced to 
where his opponent was sitting on a tall stump. Puneke made one 
great bound and although his foe stepped back, the latter movement 
was just a fraction too late as Puneke's weapon caught him under the 
jaw and killed him. Puneke married two sisters, Hinepiki and Te 
Waiata (daughters of Tuna and Marama), and by the former wife he 
had fifteen children, and by Te Waiata five, so it can be seen he did 
his part in keeping up the census returns in ancient Maoridom. 

A SOUTHERN PAKEHA-MAOUI. 

The story of James Caddell (the Pakeha-Maori) is well-known. 
In ** Muriliiku" (R. McNab) it is stated that the chief who killed the 
crew of the "Sydney Cove,*' near South Cape, was Hunneghi, and 
that Caddell married Tougghi-Touci. One of my native informants 
said, "James Caddell (better known as 'Jimmy the Maori,' because 
be was tattooed by our people, or * Jimmy the Boy,' because lie was 
so young when captured) was taken prisoner near South Cape 
(Opehia or Puhi-waero) by Maoris from Euapuke under Te Pahi. I 
will tell you the lineage of that chief. Hau-tapu-nui-o-Tu married 
Taumata and had a daughter, Te Whakaraua, and two sons, Honekai 



Digitized by 



Google 



Traditions and Legends. 159 

and Pukarehu. Te Whakaraua was an ancestress of the late ciiief 
Topi. Honekai married Kohu-wai, a Kati-Mamoe woman of liigli 
rank, and begat Kura (who was tlie mother of Tuhawaiki) and 
Whakataupnka. Tliis last chief was very ugly, and was called **01d 
Wig'' by the whalers. He was good to the Europeans, but cruel to 
his own people. His children all died, and that ended his line. 
Pukarehu married Koko and begat Te Pahi, whose only child was a 
son, Te Kalia, who was drowned with a canoeful of people in Foveaux 
Straits (Te Ara a Kiwa) and ended that branch. Te-Pahi had a 
young sister, Tokitoki, and it was she who threw a mat over ' Jimmy 
the Boy * and saved his life. She afterwards married Jimmy, and 
went over to Sydney with her husband. Te Pahi went over later, and 
both died in Parramatta, and thus ended that branch.'' 

It will be noticed that this statement bears out Rutherford's 
account about the way Caddell's life was saved, the other accounts 
stating that the boy ran to an old chief and happening to touch his 
ka-ka-how [kakahu — a garment), his pei-son was then held sacred. The 
name Hunneghi evidently stands for Honekai (an uncle of Te Pahi), 
and Tougghi-Touci for Tokitoki. This latter weird-looking mis- 
spelling is perhaps more understandable when we remember that the 
Maoris in the south often pronounce **k" as **g." 

fTo he continued. J 




Digitized by 



Google 



AN ANCIENT CARVED PARE. 



THEOUGH the courtesy of Mr. T. F. Cheesemaii, F.L.S., curator 
of the Auckland niuseuin, we are enabled to piiblisli a photo of 
a very beautiful pare^ or architrave, used by the Maoris in old days to 
ornament the doorways to their wha/re-punis, or principal houses. Mr. 
Oheeseman says, in his annual report for 1918-19, . . . " it is evidently 
of great age, and in perfect condition. It was dug up in a peat swamp 
in the Hauraki plains [Thames district] , and has been deposited for a 
lengthened period by the finder, Mr. L. Carter." 

It seems probable tliat the pare had been hidden in the swamp to 
preserve it from the enemies of the tribe to whom it belonged, and it 
may be suggested that it was during the numerous raids of the 
northern Nga-Puhi tribes into the Thames district in the early years 
of the. nineteenth century, that the pare was hidden away. Swamps 
were common hiding places for wood carvings under similar circum- 
stances, for they were always well preserved there. 

The central figure is a female ; but what it represents is difficult 
to say. We suggest it may be intended to delineate the goddess 
Hine-nui-te-po, who presides over Hades, and ** who drags mankind 
down to death,'' the smaller figures right and left of the large one 
being human spirits in the process of being drawn to **the great-lady- 
of-night," as was the hero Maui when he attempted to destroy death 
on earth, and himself fell a victim to Hine-nui-te-po. The pare is 
eight feet long, and the delicate carving, especially that to the right 
and left of the lower part of the central figure, is the best we have 
seen. 



Digitized by 



Google 




ft: 

C 
> 

< 

< 



Digitized by 



Google 



Digitized by 



Google 



TRADITIONS OF AND NOTES ON THE 
PAUMOTU OR (TUAMOTU) ISLANDS. 



Collected by the Rev, Fire Hervi Audran, of Fakahina, 
Faumotu Islands, 

TKANSLATED FROM THE FKENCH BY R. H. ROCKEL, M. A. 



PART IV. 

(Continued from Vol, XXVlll.,page 38,) 

Fakahina or Predpriatie Island, 
discovery and geographical situallon of the island. 

1. According to Bienzi tlie Island of Predpriatie was discovered ii> 
1824 by Kotzehue. He found it peopled by a vigorous, olive- 
coloured race. Pretty huts, built of brandies, were seen here and 
there under tlie trees. The swell and the attityde of the natives 
prevented him from landing. The island is low-lying and wooded, 
and has an interior lagoon. It is four miles in extent from E.N.E. ta 
W.S.W. Its geographical position, as ascertained by that navigator^ 
is 15^ 58' S. and 142° 39' W. [west of Paris] 

2. According to **Tlie Pliysical and Political Geography of the 
French Possessions of Oceania," by M. F. V. Piquenot, Fakahina or 
Fakaiiift or Predpriatie is 540 sea-miles, as the crow flies, from 
Motu-uta in the roadstead of Papeete. He notes Fakahina as an 
island that produces but little copra. This is a great mistake, as the 
whole crop might amount to some 600 tons annually. He further 
states that the area of the island is four miles by four, and that the 
population is 131 souls. All these figures are incorrect. 

3. According to the French map. No. 1716, of the Tuamotu 
Archipelago, Fakahina is situated E.N.E. of Tahiti, from which it is 
distant 580 miles in a straight line, and 660 by a detour via Hao. The 
actual landing-place of the island, situated on the east, is in lat. 16*^ 
r S., and long. 142° 29' W., near the great rock of Tenanako, which 
has given its name to the whole of tliis part of the island. 

TOPOGRAPHICAL DESCRIPTION OF THE ISLAND. 

The situation of Fakahina is incorrectly marked on maps of the 
Archipelago. Not only is this the case, but it is assigned a ft)rm 
which it does not possess at all. It is almost circular instead of being 
oval. Thus the egg-shape given it is incorrect. In form it is au 



Digitized by 



Google 



162 JOURHAL OF THE POLYKESIMtl SOCIETY. 

elongated crown almost witlnrnt indentations. Unlike most of the 
islands of the Paumotu gi-oiip, wliose 8.K coasts nsually exhibit only 
a few clumps of trees, sciittered at rare intervals, and seemingly 
given l>y Providence as a guide for captains who are in the liahit of 
sailing tliese seas, this island is planted and well wooded over the 
whole of its area. Kereteki of Fakahina — to employ the specific term 
used by the natives to designate this part of the island — has only one 
hoa (channel), which coninninicates with the open sea. With the 
exception of this hoa^ and of one or two other spots, where tlie soil 18 
poor, the whole island is planted. The south-east part of tlie island 
constitutes the fourth division for the production of copra. But, as it 
is from time to time invaded and submerged by the high tides 
prevalent during the equinoxes, its productivity is less than that of 
the other divisions. 

AREA. 

In length the island measures from east to west and from reef to 
reef, seven miles, five of which are occupied by the lagoon. Its 
breadtli from north to south is five miles, of which four are occupied 
by the lagoon. The strip of cultivated land which suiTounds the 
island varies in breadth/ From the western reef, where a landing is 
usually effected, to' the dry stone wharf constructed on the edge of the 
lagoon, measures about half a mile. At Te Matahoa the distance is 
greater. It. was at this spot, apparently, that M. Gamier, who had 
been sent to deepen and im]>rove the channels in the Tuamotu Group, 
effected the triangulation of the island. Possibly it is as a result of 
M. Gtirnier's labours that it has been ascertained that the situation of 
the island has been ** incorrectly marked on the maps.*' 

The former name of Fakahina was, according to an ancient song, 
Niuhi, ** the isle of the coconut." As a matter of fact, the word niM, 
which is pure Maori, means " coconut." The final syllable hi is only 
a suffix added to round off the word or as a ** glide " formed by the 
fall of the voice. The real derivation is possibly, rather, from the 
Polynesian »/m, ** coconut," and the Marquesan word ehi, also 
^* coconut." The final syllable of this word being suppressed, we 
have the compound word signifying **rich in coconuts." Niuhi bears 
out, to a letter, the signification of its name. It is, in fact, at tlie 
present time one of the most prosperous islands of the archipelago. 
Here more than anywhere else we can admire the beauty of tlie 
tropical scenery. The numerous two/w (islands) C(»vered with a luxuriant 
vegetation seem so many baskets of verdure cast by Providence into 
the midst *»f the Pacific;. The whole is but a veritable forest of 
coconut palms. It is so extensively planted with this fruitful tree 
that it seems to the wondering gaze a huge unbroken crown of 
exuberant verdure, surrounding its tiny lagoon, from the surface of 



Digitized by 



Google 



Traditions of the Paumotu Islands. 163 

which rise gracefully several islets. These are veritable clumps of 
dark mikitniki of a velvety emerald -green edged by the strip of sand 
sparkling in the first rays of the rising sun. 

The name Fakahina, according to the elders, is comparatively 
recent ; and the author and cause of the cliange are quite unknown 
to-day. In my opinion the name is of Marquesan origin. If tlie 
Marquesan Dictionary of Mgr. Dordillon ])e consulted, the following 
derivation may be made out; faka in Polynesian signifies "to make," 
and hina signifies conquered, vanquished, overthrown, defeated. The 
word hina, pronounced hinya signifies in Tuamotuan the same thing, 
and has exactly the same sense. 

The following is a song describing Niuhi, i.e., Fakahina: — 
Ko Niuhi te fenua tukau matagi, This is the land of Niuhi, 
tere te vaka i tua kiriti hia where blows the gale so 

tere te vaka i aro kiriti hia strongly, that when a canoe 

kia hipa hia tura e koe sails in the offing it is driven 

te tika ia o te henna o Niuhi out to sea, and also when 

kua taka pipita raufara sailing homewards it is 

ka to ilunho driven out to sea. 

ka to gahegahe W<mld you mark the form of 

fakarua te matagi Niuhi you will see that it is 

paupau taku manava. like a roll of raufara (pan- 

K horote kaiga Niuhi te fenua i te danus leaves rolled on a 
fatatapua Nuku o stake) at the hour when day- 

rarahiva light dies, and a great calm 

Hiva tautua fills the broad horizon, a 

calm broken from time to 
time by light breathings of 
wind from the north-east. 
But I can sing no more ; my 
}»reath is failing. 
fair land of Niuhi ! 

(Hiva tauaro, western portion of the Taumotu Group; Hiva 
tautua, eastern portion.) 

In this ancient lay we are told in clear and imnressive terras that 
Niuhi is a land of violent and irresistible storms, which cany (mt to 
sea })oth canoes which are making for the island and those which are 
leaving it. But when the storm is over and perfect calm reigns, and 
when the gentle breeze from the north-etist breathes in the 
brilliant sunshine, so round does Niuhi appear tliat it bears a striking 
resemblance to a roll of pandanus leaves. 

ORIGIN OF THE FIRST MAN AT FAKAHINA. 

With respect to the first man wlu* inhabited Fakahina, I have been 
unable to gather any information at all definite. That the first 



Digitized by 



Google 



164 JOURHML OF THE POLYNESIAN SOCIETY, 

inliabitaiits were iinmerous is evident from the fact that several old 
met), when questioned on the point, quoted tlie names Tane-tupu-hoe, 
Maliinui, Maui and Mapu-te-agiagi, Te-mapu, Te-agiagi (first 
migration.) 

Are these people autochthones? It is difficult to believe this. In 
my opinion all the tribes of the Tuamotu Group are immigrants. In 
the case of tlie inhabitants of Fakahina, as of many others, tradition 
preserves a complete, or nearly complete, silence. Of their original 
home, the cause of their coming, the name of their pahi (vessel), their 
wanderings by the way, the time of their arrival there is not a word. 
In a word, all is uncertainty with regard to the first inhabitants of 
Fakahina and their origin. 

There are, however, two different opinions. It is claimed by some 
authorities, but without any proof in support, that the first inhabitants 
came from the west, that is, from Tahiti, or, at least, from the islands 
near Tahiti. Othei*s, with more probability, claim that they came 
from Mangareva. Te-mapu Te-agiagi, one of the very first inhabitants 
of Fakahina, was certainly a native of Mangareva. He was, no doubt, a 
son or grandson of Te-agiagi, high-priest of Gatavake, youngest son of 
Anua-motua, who came from the Sandwich Islands. 

The gi*eat migration which peopled Beao, Pukarua, Takoto, 
Vahitahi, Hao, Fakahina, Fagatau, and, to some extent, Hikueru, by 
giving them wives, took place iu the reign of the celebrated Ape-iti 
XXV., king at Kikitea at Mangareva, conqueror of Taku. From 
Ape-iti to Gregorio Maputeoa, the last king of the Gambler Islands, 
who died on June 20th, 1857, there was a succession of kings, ten in 
number, who reigned in Rikitea. Now, if an average of twenty years 
be allowed for each reign, we should have about 220 plus 60 years, a 
period which would yield tlie year 1637 as the date of this great 
migration. In a war wliich broke out between the districts of Rikitea 
and of Taku at Mangareva the latter was victorious. Rather than 
submit and accept the conditions imposed by Ape-iti the former people 
preferred to migrate elsewhere and preserve unimpaired the freedom 
which was their dearest possession. The then reigning sovereign of 
Taku was Tupou. He was conquered by Ape-iti, and almost all his 
tribesmen slain. He took refuge with such of his people as survived 
in the Tuamotu Group. They escaped on seven rafts. (SeeMangarevan- 
French Dictionar}-, p 22, s.v. Ape-iti, and this "Journal," Vol. XXVII., 
p. 130.) Some years ago (1883) an old man of Taravai informed the 
present chief of Fakahina, Bernard Mahui, that the fii*st inhabitants of 
Niuhi came from Mangareva. The reason for this migration was, 
according to this old man, a bloody war iu which they had been 
conquered. They had no alternative but to flee or be taken, slain and 
sent to the ovens. 

A further fact, more recent perhaps, is that the elders of Fakahina 



Digitized by 



Google 



Traditions of the Paumotu Islands. 165 

understood aud spoke perfectly the Maugarevan dialect. The 
following incident is a proof of this : In the days of paganism at an 
unknown date, but many years before the aiTival of the Catholic 
missionaries, a Mangarevan pahi^ or canoe, put in, and the crew 
landed. In the island everyone was on the alert and expecting war. 
The new-comers were asked if they had landed as friends or enemies. 
Meanwhile both sides made active preparations and had already made 
their dispositions for battle. Just as a bloody strife was about to 
commence challenges in good Mangarevan were made from camp to 
camp. Tliis was sufficient to calm those spirits which were already too 
eager for the combat. It may be remarked that to land without a 
safe-conduct on the reefs of one of these islands, where cannibalism 
was still rife, was to risk being sent to the oven. No mercy was shown 
to a foe. Therefore, before landing, a crew had to hail from a distance, 
that is, from the canoe, the inhabitants who were standing ready with 
their arms on the reef. It was fii'st necessary to establish one's parau- 
tupuna, that is, one's genealogical tree so as to find in the island an old 
relative, some ancestor, who ipso facto became a protector and a 
guardian. As soon as the relationship was established by the inhabi- 
tants of the island, the visitors could land without danger. They were 
feiiij or relatives, and received as such. The voice of blood was always 
respected. 

HAS FAKAHINA BEEN LONG INHABITED? 

I have investigated this question, employing the genealogical 
records of several families, who have willingly placed them at my 
disposal and permitted me to prosecute my researches, although among 
our natives this is a great and carefully guarded secret. I have 
examined and carefully studied seven different genealogical tables. 

•The longest of these, that which begins from Tane-paku and Kai- 
haruma,his wife, counts only twenty-tiiree generations. Now, reckoning, 
on an average of thirty years for a generation, as is done in Europe, 
we should have 690 years, bringing us back to the year 1227.* The 
genealogy of tlie Gati-Mahinui, wliich begins with Panoko and his wife 
Kulii, reckons fifteen generations. This would give us 450 years and 
bring us to the year 1467. Tliat beginning with Marere and his wife 
Te-Pogi, contains from twelve to fifteen generations, brings us to 1467 
or 1557, according to the line of descent followed. From Marere and 
his wife Manuia, are descended nineteen generations, giving 580 years 
and bringing us to the year 1347. The line of descent which claims 



* The PolyneHian Society, after ascertaining the views of those most capable of 
judging, decided from the very first to adopt 25 years as the length of a Polynesian 
generation; and P^re Audran^s ought to be altered to ag^ee with this scale, and 
thus allow of comparison with dates in other groups. — Editob. 



Digitized by 



Google 



166 JOURKML OF THE POLYtlESIMIf SOCIETY. 

Manava-rere-tu-te-uiiu for tupuna (ancestor) reckons only ten genera- 
tions. No donbt it was this Manava-rere wlio landed on the island of 
old,' and the whole of whose crew was surrounded, captured and slain ; 
but it is not said tiiat Manava-rere liiniself was killed. Evidently he 
alone was spared, and afterwards became the founder of a branch of 
the race. The parau tupuna of Tane-niatavai and his wife Kapu-roro 
reckons twelve generations down to the children of J. Tagaroa and 
R. Hura, giving 360 years and bringing us to 1557. The parau tupuna 
of Rata, tliat is, beginning from Kulii and Panoko down to their 
descendants Pakora and Haroagi, reckons eight generations and brings 
to the year 1677. From Rutua and his wife Te-Fakaruru have sprung 
ten generations. 

According to an ancient and persistent tradition it was Teliu, the 
son of Te-Tahoa and Te- Ahio, wlio on one of his voyages to Taliiti or 
to some of the islands to the west, in his famous pahi, '^ Katau," 
brought to Fakahina the first coconuts, taro aud ape. On tlie shore of 
the island lie lost his anclior, a large, flat, round stone resembling a 
mill-stone, pierced with a hole in the middle. This stone encrusted 
with coral is still visible. From Tehu have spmng six generations 
extending, according to our average, over a period of 180 years.f 

It may be said that the coconut, which covers the greater part of 
the cultivable area of the island is the chief source of the wealth of the 
whole archipelago. The copra from the Tuamotu Group is of much 
superior quality to that produced in Tahiti, and commands a higher 
price. It is owing to the quality of tlie former that Tahitian copra is 
so well-known and so eagerly sought after. Some 9,000 tons, valued 
at 4,400,000 francs have been placed on the European market. Of this 
quantity the island of Fakaliina furnished a large proportion. Tlie 
following table is given for the purposes ot information : — 

Years .. 1912 1913 1914 1915 1916 
Tons .. 170 490 296 400 395 

These are the official figures for the island, but in reality they are 
too low. Of a certain am(»unt no official record is taken. The chief or 
his deputy is quite unable to supervise all the weights and exports. 
Again, a certain amount of evasion goes on, the supercargoes giving 
whatever figures they please. The fact is well-known and cannot be 
disproved. The island is one of the most prosperous in the archipelago 
at tlie present moment. Considering its small area it surely holds the 
record for an annual yield. It is possible to make tlie circuit of 
Fakahina in less than a day at a leisurely walk. 



t We omit here, for want of space, the author's description of the uses to 
which the coconut is applied, as they have often been described in the pagfes of this 
** Journal."— Editor. 



Digitized by 



Google 



Traditions of the Paumotu hiands. 167 

The following is a list of the different coconuts found in the 
Tuamotu Group : — 

1. The Mamagu : — Tlie fibre is dark green, almost black ; th& 

kernel wlien cut dries very quickly. 

2. The Kaipoa : — The husk is so tender aud sweet that it is eateik 

whole. The natives are very fond of it, especially tlie children, 
who are always nibbling at it or sucking it. There are the 
ordinary Kaipoa and Maki re -Kaipoa, that is, the i)ne whiclj 
yields enormous crops. From it a salad can be prepared. 

3. The Uakuru: — ^This is a variety of Makire, that is, a coconut 

yielding an abundance of fruit.* 

4. The Takaveatika : — In this the spathes are not branched, the 

fruit being closely packed and usually attached directly to the 
bough itself. 

5. The Koheko:— Tlie fruit is almost yellow. 

6. The Uriuri : — Tlie fruit is almost as green as that of the Mamagu» 

7. The Fafatea. 

The coconuts whose stripes are very long and very higli are called 
Hahari a Eva, ' 

I estimate that, at the lowest computation, tliere must be 48,000 
coconut trees on the island. My estimate is based on the following 
calculation. Eighty coconut trees are necessary to produce annually 
one ton of copra. Now, Fakahina, I feel sure, can produce 600 tons- 
annually. This gives 48,000 trees, as above. If we take an annual 
yield of 360 tons, we should get 28,000 trees; but in 1913 the yield 
was 490 tons. Many coconuts are eaten by the inhabitants or given to 
the pigs. I have seen hundreds of folk dressed in silk (?) and well skilled 
in the husking of the nuts and exposing them to the sun, end up b^ 
breaking the nut and then eating it. At least half a ton of copra 
disappears in this way in a single day. Moreover, there are thousands- 
of coconut tr^es which never bear a nut the whole year round, as they 
are smothered by the larger trees. 

For the purposes of the coconut harvest the island is divided into 
four rahui, or sectors. At the entrance of each sector the established 
rule, accepted by everyone, is to remove the husk carefully before 
cutting the nut and making copra. 

Immediately on their return from Mangareva, the people of 
Fakahina planted only the interior of their matte, Tliey vied with 
each other in planting the largest number of trees, and in doing so the 
most rapidly. Consequently, the plantations of this period are lacking 
in symmetry, and are planted too thickly. In their ignorance many 
natives are persuaded that the more trees there are the better. In 
many of the islands it is still very difficult to remove this prejudice, 
and to make the natives understand that it is in their own interests, 
and til at of the community, to allow sufficient space between tlie trees* 

fl'o he continued. ) 

Digitized by VjOOQIC 



POLYNESIAN LINGUISTICS. 



IV.— POLYNESIAN LANGUAGES OF THE SANTA 
CBUZ ARCHIPELAGO. 



By Sidney H. Ray, m.a., f.r.a.i. 



1. Introduction. 

2. The Languages of the Santa Cruz Archipelago. 

3. A Pileni Grammar. 

4. A note on the Tikopia Language. 

6. A Vocabulary on the Pileni Language. 
6. Pileni Texts. 



I. INTRODUCTION. 

THE Santa Cruz Archipelago consists of four groups of islands in 
the Western Pacific, lying between 9° and 12° south latitude 
and between 165® and 171° east longitude. The south islands of the 
Solomons are due east from Santa Cruz, the Torres and Banks 
Islands lie to the south, whilst far away east is Rotuma, and in the 
south-east, Fiji. • 

The four groups are : — 

1. The Duff or Wilson Group in the north, consisting of 

Taumako and several smaller islands. 

2. The Reef or :Swallow Islands in the north-west, including 

the islands of Nibanga, Banepi, Lomlom or Fonofono, 
Fenua Loa, Matema, Nufiloli, Pileni, Nukapu, Nalogo 
and Nupani, with the island and active volcano of 
Tinakula about twenty-five miles south-west of Matema. 

3. The large island of Santa Cruz (Ndeni or Indenni) about 25 

miles long and from 10 to 12 miles broad, with the smaller 
islands of Trevanion close to its north-western corner, and 
Lord Howe Island off the soutli-east coast. The native 
name of the northern side of Trevanion Island is Te Motu, 
that of the southern side is Nimbi. 



Digitized by 



Google 



Polynesian Linguisfies. 169 

With these may also be included the islands of Utupua^ 
about forty miles south of Santa Cruz Island and Yanikoro^ 
or Yanikolo, about 25 miles south of Utupua. 
4. Nearly 150 miles south-east of Vanikoro is Tikopia, with 
Cherry Island (Anuta) about 80 miles to the north-east^ 
and Mitre Island (Fataka) about the same distance east* 
north-east. 

Santa Cruz was discovered by the Spanish navigator Alvaro de 
Mendana on his second voyage to tlie Pacific in 1595, in search of 
the Islands of Solomon, which he had discovered on his first voyage 
in 1567. Having lost his bearings in tliick weather, one of his ships,, 
the ** Almiranta," disappeared off the island of Tinakula during the 
night of September 7th, and next morning tlie mainland of Santa 
Cruz was in sight. Tliis was at first thought to be one of the Solomon 
Islands, but when the mistake was discovered Mendana named the 
new land Santa Cruz. He tried to establish a colony, but a mutiny 
broke out amongst his crew, and the murder of a friendly chief,. 
Malope, by the mutineers aroused the hostility of the natives. 
Mendana died of fever, and his lieutenant Quiros led the expedition 
back to the Philippines. 

In 1605 Quiros revisited the Pacific and discovered Taumako. 
From the chief of this island, named Tamay, he obtained information 
of Chicayana (i.e. Sikaiana) and other islands in the west. Among- 
these were Fonofono (Lomlom), Nupan (Nupani), and Pilen (Pileni) 
in the Reef Islands ; Tucopia (Tikopia) and Manicolo (Vanikolo).^ 

In 1767 Carteret in the ** Swallow " rediscovered Santa Cruz. He 
visited the Reef Islands and named them after his ;sliip. Thirty-one 
years later Captain Wilson, conveying the first Protestant missionaries 
to Tahiti, passed Taumako and gave the name of his ship the **Duff '*^ 
to the Group, to which it belongs. 

In 1785 the French frigates "Astrolabe" and "Boussole," under 
the command of La Perouse, left Brest on a voyage of discovery in the 
Pacific. They arrived at Botany Bay in January 1788, and left in 
February. Then for nearly forty years their fate was unknown, till in 
1827 Captain Peter Dillon, of the Hon. Bast India Company's* ship 
** Research,** ascertained that the ships and their crews had been lost 
off Vanikoro. Meanwhile Captain Edwards in 1791, in the "Pandora,'^ 
had passed between Utupua and Vanikoro and discovered Cherry and 
Mitre Islands. The French government sent an expedition in search 
of La P6rouse in 1791, in the ships "Recherche" and "Esperance," 
under the command of General D*entrecasteaux. In 1793 these ships 
called at Santa Cruz. Mr. Woodford notes that "in one of the 

1. Of. " Journal Polynesian Society," XXI., 1912, p. 167, and XXVI., 1917^ 
pp. 34, 35. 



Digitized by 



Google 



170 JOURNAL OF THE POLYNESIAN SOCIETY. 

canoes which came alongside a portion of a joiner's chisel fitted to a 
wooden handle, was observed. It is possible, and even probable that 
this chisel came from one of La Perouse's ships, since the distance 
from Santa Cruz to Vunikoro is only sixty ndles, and communication 
by canoes is not infrequent, but the clue was not followed up.'** 
Dillon first came upon relics of La P6rouse in 1813 at Tikopia, but the 
fate of the expedition was not fully established till 1827.^ Confirmation 
was obtained by Dumont d'Urville in 1828. The same navigator 
revisited the islands ten years later. 

• The modern history of the Santa Cruz Archipelago began with 
the first visit of Bishop G. A. Selwyn in 1852. He did not then go 
ashore, but four years later tried at Santa Cruz to make friends with 
the people. His knowledge of Maori enabled him to make himself 
understood a little at Nukapu in the Reefs. He also landed at Utupua 
and Vanikolo. 

In 1862 Patteson, then Bishop of Melanesia, went ashore at Santa 
Cruz at seven different places. In 1864 two members of the 
Melanesian Mission, Edwin Nobbs and Fisher Young, were killed at 
Oraciosa Bay, Santa Cruz. Patteson revisited Nukapu in 1 870, and 
again in 1871 when he, the Eev. J. Atkin, and a native named 
Stephen were killed. In 1875 Commodore Goodenough was killed 
at Carlisle Bay, Santa Cruz. The taking home of some Nutiloli men 
who had been cast ashore in the Solomons in 1877, led to the 
re-opening of the mission, Wadrokal, a Loyalty islander, being the 
first teacher. Since that time the attitude of the islanders to white 
men has sensibly changed, and Christianity has^made some progress 
in Pileni, Nufiloli, and Santa Cruz. 

The anthropology of the people of these islands has been partially 
dealt with byGaimard'* and Dr. Codrington,* and short but vivid 
sketches of native life have been given by the Rev. W. O'Ferrall.^ 
The sociology of Santa Cruz, Vanikolo, the Reef Islands and Tikopia 
has been discussed by Dr. Rivers, ^ that of Tikopia and Santa Cruz 

2. Handbook of the British Solomon iHlands Protectorate. Tulagi, British 
Solomon Islands, 1911, pp 11-13 

3. Of. P. Dillon. ** Narrative and succeHsful result of a voyage in the South 
Sea8," London, 1829. 

4. Gkiimard. ** Voyage au tour du Monde.'* Paris, 1833. Tome V., p. 108 
fP, and pp. 304-312. 

5. R. H. Oodrington. *'The Melanesians: Studies in their Anthropology 
and Folk-lore." Oxford, 1891. 

6. W. C. O'Ferrall. "Saufca Cruz and the Reef Islands." Mehmesian 
Mission. 

7. W. H. R. Rivers. **Thc History of Melanesian Society.*' Cambridge, 
1914. pp. 217-231 and 298-362. 



Digitized by 



Google 



Polynesian Linguistics. 171 

more briefly by Rev. W. J. Durrad® atidW. Joest* The island trade 
1*8 dealt with by Thileiiius. * <^ Graebiier has published a long account 
of the Santa Cruz people. ^^ Tlie **Folk-Lore " has been exhibited by 
Eev. W. 0'FerralP2 a„(| j)^ Codrington.^ ^ 

III physical features the Santa Cruz islanders are Melanesian, but 
in Tauniako and some of the Reef Islands tliere appears a strong 
strain of Polynesian blood, whilst in Tikopia the people are Polynesian 
witli traces of Melanesian admixture. Additional evidence of this 
mingling of races is found in the language, and in the freer inter- 
<5ourse of tlie sexes. The natives of all the islands chew betel. Kava 
is only used in ceremonies in Vanikolo, IJtnpua and Tikopia, and not 
At all in Santa Cruz and the Reef Islands. The canoes, some of very 
large size, are dug-outs witli outriggers, and not plank-built as in the 
Solomons. The Taumako people are said to be the best canoe 
builders, and the Matema people the best sailors. 

Native money (Javan) is made from small red feathers. Mats are 
woven on a loom in Santa Cruz and Nufiloli, but not in Vanikolo, 
TJtupua or the Reef Islands. The fibre is obtained from the stem of 
the banana, a black-stemmed variety being used in forming a pattern. 
Weaving was seen by the Spaniards in 1595, and is no doubt of 
ancient use. ** Native tradition is that it was invented by a woman, 
but insomuch tliat it kept lier from the less interesting and more 
arduous labour in the gardens, the invention was appropriated by her 
lord and master."^ "* Tlnmgh almost identical in form^-^ with the loom 
used in the Polynesian Solomon Islands and in the Carolines, there is 
no evidence of its introduction from those places. The loom and its 
parts have each tlieir distinctive names in Santa Cruz and Nufiloli, 
but these are entirely different from the names in the Solomon and 
Caroline Islands. The Polynesians in the Reef Islands have no looms 
and can only describe the apparatus and its parts in common terms as 
*' sticks " or ** strings." 

8. Rev. W. J. Durrad. '* Southern CroHs Log " 19. 

9. A. Biiessler. *' Neue Sudsee Bilder," 1900, p. 386. 

10. Dr. G. Tliileiiius. ** EthuographiHche Ergebniese aus MelaneBien ** I. 
Halle, 1902. 

11. F. Graebiier. '* Volkerkunde der Santa Cruz Inseln. Ethnologica.'* 
{Stadt-RautenKtrauch-Joest-Museum.) Kijln I., 1909, pp. 71-184. 

12. Rev. W. O'FeiTall. *' Native Stories from Santa Cruz and Reef Islands," 
^♦Journal Anthropological Institute," XXXIV., 1904, pp. 223-233. 

13. R. H. Codrington. '* Melanesian Folk-tales. Folk-lore," IV., 1893, 
pp. 509-511. 

14. O'Ferrall. " Santa Cruz and the Reef Islands. Mat making." 

15. Cf. pictures in O'Ferrall and Parkinson, ** Dreissig Jahre in der Sudsee." 
Stuttgardt, 1907. 



Digitized by 



Google 



172 JQURHAL OF THE POLYKESIAK SOCIETY. 

The distinction between chiefs and commoners is very definite in 
Tikopia, ^ ® but in Santa Cruz the authority of the village headmen 
" does not extend beyond that village, though their influence may do 
so," and there are "no powerful chiefs such as are found in the 
Solomons."^ ^ 

Each village in the group has its club house, (Santa Cruz, Mandai ; 
Vanikolo, Manggore ; Nufiloli, Japulau ; Pileni, Afalau) in which 
the unmarried men live, and where male visitors from other villages 
are entertained. Each has also its "ghost-house,'' (Santa Cruz, 
Ma-ndaka; Nufiloli, Niekipejalikive; Pileni, Faleatua **) containing 
wooden stocks or posts, " ghosts " (Santa Cruz, Nduka ; NufiloL", 
Ndekilavo ; Utupua, Nduo ; Yanikoro, Lenoe ; Pileni, Atua) carved 
and painted to represent a dead person who in his lifetime possessed 
great power, influence or skill. (Santa Cruz, Malete ; Nufiloli, 
Kinaa ; Utupua, Ana ; Pileni, Mana. ^ ** ) 

After death Santa Cruz spirits go to Mlduka,^® those of Nufiloli 
to Patanee, those of Pileni to Thalafali. These names are equivalent 
to the Mota and Banks' Islands Panoi. 

n. THE LANGUAGES OF THE SANTA CRUZ ARCHIPELAGO. 

The first word recorded in the language of this region of the 
Pacific was obtained at Santa Cruz by Mendana. Figueroa ^ ^ states 
that the chief Malope, who came on board, called himself Tarique, a 
word which is, no doubt^ intended for the Polynesian Te Ariki. Its 
use on Santa Cruz, where the language is M^elanesian, may be 
accounted for by the fact that the Spaniards in order to obtain 
interpreters had sent to kidnap some boys from the Beef Islands. 
These were said to be more intelligent than the people of Santa Cruz, 
a supposition confirmed by native admissions to Dr. Codiington.^^ 
Thus, instead of tlie Santa Cruz word for chief, Mbonie, the Spaniards 
obtained the Reef Island interpreter's Te Ariki or Te Aliki. Gaimard 
gives a form of Te Ariki used in Tikopia. During Dumont d'Urville's 

16. W. H. Rivers. " The History of Melanesian Society," Cambridge, 1914, 
p. 305 ; and Rev. W. J. Durrad, " Southern Cross Log,** 1911. 

17. Rev. W. O'Ferrall. ** Santa Cruz and the Reef Islands.*' Introduction. 

18. Dumont D*Urville gives ** Baito Atoua, Baito Tapon, Maisondes esprits," 
op. cit., p. 169, but these words are in the language of Tikopia, and show that he 
probably had a Tikopian interpreter. The Vanikoro equivalent would be 
Moelefioe. 

19. Cf. W. H. R. Rivers, op. cit., I., p. 230. 

20. According to Dr. Codrington, ''Melauesians,** p. 264, the dead assemble 
at a place called Netepapa and thence go to Tamami, the volcano Tinakula. 

21. Cf. Dalrymple. "An Historical Collection of Voyages.** London^ 
1770-71. I., p. 81. 

22. Islands of Melanesia. R. H. Codrington, D.D. ** Scottish Geographical 
Magazine,** 1889. 



Digitized by 



Google 



Polynesian Linguistics. 173 

visit to the group, his naturalist, M. Gaimard, collected vocabularies 
of three dialects of Vanikoro (Vanikoro, Tanema, Taneanou) and of 
Tikopia. He also obtained the numerals of Utupua (two dialects), 
Fonofono (Lonilom), Mami (probably a village in [omitted by Mr. 
Ray]), and Indeni (Santa Oruz).^^ A discussion of the affinities of 
Vanikoro, based on these, was made by Lesson** and D'Eichthal.'* 

Inglis obtained the numerals of two dialects of '* Vanikolo, Queen 
Charlotte's Island."*® The first of these is Polynesian, resembling 
[omitted by Mr. Ray], and the second resembles the Tanema of 
Gaimard. 

A short vocabulary of Vanikoro is to be found in the Journal of 
Commodore Goodenough, with one word from Santa Cruz.*^ 

Dr. Oodrington has published short grammars and vocabularies of 
Santa Cruz (Nelua on the central north coast) and of Nifiloli (Nufiloli) 
in the Reef Islands.*® The Rev. 0. E. Fox has published longer 
vocabularies of Santa Cruz (Te Motn on the north coast of Trevanion 
Island) and Vanikoro.*® The Rev. W. J. Durrad's vocabulary of 
Tikopia, edited by Archdeacon H. W. Williams, has been published 
in the "Journal" of this Society. ^^ 

A few books in the Santa Cruz language, and one in Pileni, are 
used in the Melanesian Mission. 

Some of the languages in this region are Polynesian and some 
Melanesian. A Polynesian language is spoken in Taumako, and the 
same language is spoken in the Reef Islands Fenua Loa, Pileni, 
Matema, Nukapu and Nupani. The language of Tikopia is also 
Polynesian, but differs slightly from the Pileni. Tlie statement made 
by Thilenius that the language of Utupua is Polynesian is in- 
correct. ^ ^ 

The remaining languages of the group, so far as thej*^ are known, 
appear to be Melanesian of a very distinct type, differing a good deal 
from one another and from the Melanesian of the Solomon Islands 

23. Voyage de r Astrolabe. Philologie. Tome II. Paris, 1883. pp. 165-174. 

24. P. A. Lesson. ** Vanikoro et ses habitants.** Revue d*Anthropologie V., 
1876, pp. 262-272. 

25. Not g^ven by Mr. Ray. 

26. Report of a Mission Tonr in the New Hebrides. By Rev. John Inglis. 
Auckland, 1871. Also in *' Journal Ethnological Society " III. 1854. 

27. Journal of Commodore Goodenough. 

28. *' Melanesian Languages.*' Oxford, 1885. pp. 39-52 and 486-498. 

29. "Vocabularies of Santa Cruz and Vanikolo.*' ** Melanesian Mission 
Press,** 1908. 

30. ** A Tikopia Vocabulary." ** Journal Polynesian Society ** XXII., pp. 
86-95 and 141-148. 

31. Ethnographische Ergebnisse aus Melanesien. I. Die Polynesischen 
Inselr., HaUe 1902, p. 23. ** In Sikaiana beschrieb mir ehi Eingeborener die 
Inseln Tikopia, Liueniua, Taguu, Utupua kurz dahin, all same men, one talk.** 



Digitized by 



Google 



174 



JOURNAL OF THE POLYKESIAK SOCIETY. 



and New Hebrides. Words are difficult to write because of the 
uncertainty and shifting nature of the sounds, wliich give the 
appearance of great differences in the vocabularies. Words are often 
quite unlike tlieir equivalents in other Melanesian languages, and the 
grammars also differ considerably. But nothing appears which would 
definitely mark any of the languages as other tlian Melanesian in 
structure. 

The following vocabulary of all the languages known to me will 
exemplify their great variety. The numerals which follow will show 
some of the difficulties experienced in recording the sounds. 

For comparison 1 add the nearest language of the Solomon 
Islands (Wango, San Gvistoval), the nearest Melanesian to the south 
(Tegua, Torres Islands), the Rotuma almost due east of Tikopia, and 
as representatives of Polynesian, Tonga and Maori. 



COMPARATIVE VOCABULARY. 



ENGLISH 


1. TEMOTU 


2. NELUA 


3. myriLOLi 


4. VANIKOLO 5. TANEMA 


Areca nut 


Kavla 


— 


— 


Namue Boia 


Arm 


M« 


Mu 


(Niniea) 


Me Menini 


Arrow 


Nipna 


(Nipna) 


iV^iepa 


Puri Punene 
(Abione) 


Banana 


Mb^pi 


Mbepi 


Noi 


Pu;ia Unra 


Belly 


Nela 


Mbole 


NU08« 


Jafiane Jenana 


Bow 


N^t^v 


Netevu 


/iaprt 


Ore Ora 

(Te^iasau) 


Chief 


Bonie? 


(Boiii^) 


— 


Teligi, 

Aligt Taligi 


Ear 


NrfetM 


Ndole 


Nugo« 


|Ma*ale,« Jj^ 
Mambaheuhi ) 


Eye 


Ma 


Ma 


(Nimbaeai) 


Mala Maleo 


Fire 


Ni^ 


Nie 


me 


Nebie Nava 


FiHh 


No 


No 


(Si) 


Nieue Ane 


Hair 


Nu«i-nae 


Ni«i-nave 


Niluu 


(Wienbaja ly, j^. 
I Wenbafianili / * 


Head 


Na^ 


Vo (Na^f) 


(Nuota) 


Baja Baja 


Leg 


No 


(No) 


Nike 


Kele Aleniui 


Man 


iS'Jjpala 


Nq)ala 


Sime 


Lamoka Ranuka 

(Kanaka) 


Moon 


Tenia 


Tema 


Nepe 


Mele Malaula 


Nose 


Notu 


No 


Notrt 


iS^ele(Nole) Nele 


Koad 


Nati 


Naji 


(Numbatage) Nene Nana 


Sand 


Ndaiio 


Ndnno 


Nikewiano, 
Numbo 


Uteka Ouolo 


Stone 


Apia 


(Alpa) 


iViva 


Vaka Vaka 


Sugar-cane 


NhIu 


— 


(Nhwu) 


Tolo Rova 


Sun 


Nh;/h 


Na;ia 


Nale 


Woie(Noie) Woia 


Tougue 


Nalapu*^ 


Lapu 


Nalinibia 


Mea Mia 


Tooth 


Nu;/i 


Niwe 


Xuotonde 


Ufie Kole 


WHt-er 


Luwe 


Luwe 


(Nuei) 


Wire Nira 


Woman 


Ovla 


Ovla 


Si//eda 


Venimo Vanime 



Digitized by 



Google 



Polyn^^ian Linguitiio^ 



175 



6. TAKXAirU 7. VAHIXOLO 8. VTJnVh. 9. WAXQO 



10. 



Areca-nut 


Boaka 


FeMer« 


— 


fiua 


— • 


Arm 


Maini 


Jfe 


Namana 


Rama 


Pini 


Arrow 


Puene 


Pure 


Kotho, 

Vogono 


— 


Liw^ 


Banana 


Unro 


Puiia 


Wuo 


Hugi 


V^tal 


BeUy 


Jaene 


Sa 


Mbasela 


Aha 


Tokw^ 


Bow 


Vijane 


Ore 


Katomono 


Ba'e 


Nu 


Chief 


Teligi 


— 


— 


Maeraha 


— 


Ear 


Tafiaini 


Mambelen, 
Tama 


Si«o 


Karina. 


Yldieime 


Eye 


Mataeo 


Mala 


Ndona 


Maa 


Mete 


Fire 


laua 


Nepie 


Nivio 


Eu 


Ntfv 


Fish 


Namokho 


Namaga 


Nono 


I 'a 


Nig. 


Hair 


Vien-Baja 


WieNe 
Mbaza 


Niviniola 


Wareha 


Ul 


Head 


Baja 


MbHza 


Jinio, 
Ndiuio 


Ba'u 


Kwutu 


Le^^ 


Aeleda 


Kele 


Nagana 


Uwa 


Rowo 


Man 


Amualigo 


Lamuga 


Tumuo 


Noui 


Tela 


Moot! 


Metele 


Mele 


Jolo 


Hura 


Mflgag<? 


No.se 


JVeleo 


iVele 


Nomboouni- 
vla 


Barisu 


Mundu 


Road 


Anaoko 


Nene 


Nduinene 


Tara 


M^'aU 


Sand 


Ouele 


Wutega 


Nene 


— 


Nin 


Stone 


Vijiioko 


Vaga 


Foio 


Hau 


Vot 


Sugar-cane 


Ton 


Lenibie 


Torio 


Ohu 


Te 


Sun 


Aeve 


Woie 


Nuo 


Sina 


Ek) 


Tongue 


Miniiaeo, 
Miniako 


Mea 


Mamela 


Meamea 


— 


Tooth 


I/e 


Wufie 


Uju 


Riho 


Liu 


Water 


Ero 


Wire 


Nio 


Wai 


Pe 


Woman 


Viwiyi 


Neme 


Aimio 


Urao 


Lekwrtvin^ 


EHOLISH 


11. BOTUMl. 


'.12. PILENI 


13. TIZOPIA 


14. TONGA 


15. ICAOBI 


Areca-nut 


— 


Pua 


(Kaura) 


— 


— 


Arm 


Siu 


Lima 


Lima 


Nima 


Rifia 


Arrow 


Kas 


jVau 


Fana^JVasaa JVahau 


Pere 


Banana 


Per 


Tavel 


Futi 


Fuji 


— - 


BeUy 


Efi 


Manava 


Afira»a 


Kete 


Kopu 


Bow 


Fan 


Loku 


Fana-kalolo 


Kau-fana 


Kopere 


Chief 


i\ra/mja 


Aliki 


Ariki, Aliki 


Eiki 


Rawatira 


Ear 


Falina 


Taliwa 


Tariwa 


Teliwa 


Tariwa 


Eye 


Maf 


Mata 


Kafimata 


Mata 


Kanohi 


Fire 


Roh 


Keu 


Afi 


Afi 


Ahi 


Fish 


I'a 


Ika 


Ika 


Ika 


Ika 


Hair 


Lera 


Laulu 


Rauru, 

Raulu 


Louulu 


Makawe 


Head 


FUo 


Piso-ulu 


Poko-ulu 


Ulu 


Upoko 


Leg 


La 


Vae 


Vae 


Vae 


Waewae 


Man 


Fa 


Tai, Tamita 


. Tawata 


Tawata 


Tanata 


Moon 


Hula 


Akafu 


Marama 


Mahina 


Marama 



Digitized by 



Google 



176 



JOURKAL OF THE POLYHESIAK SOCIETY. 



ENGLISH 


IL BOTUXA. 12. PILBNI 


13, TIXOEIA 14. TOWOik 


15. XAOBI 


Nose 


Isu lu 




Isu, Kauiau Ihu 


Ihu 


Road 


Sala Ala 




Ara 


Hala 


Ara 


Sand 


Fanfan One 




One 


Oneone 


Onepu 


Stone 


Hof Fatu 




Fatu 


Maka 


Eowhatu 


Sugar-cane 


— Tolo 




Toro 


To 


— 


Sun 


Asta Vela 




Ra 


Laa 


Ra 


Tongue 


Alele Alelo 




Alelo 


Elelo 


Arero 


Tooth 


Al Nifo 




Nifo 


Nifo 


Niho 


l^ater 


TauM Vai 




Vai 


Vai 


Wai 


Woman 


Hone Fafine 


Fafine 


Fefine 


Wahine 






NUMEEALS. 






1. TBMOTU 2. NBLUl. 


2l. indbni 


3. NUFILOLI \ 


Ja. fomovono 


1. Eja 


Eja 


Teja (Beja) 


Ni^ 


Ni^ 


12. ' Li, AU 


AU 


Ali (Odi) 


Lilu 


Lelu 


3. Tu 


Atfi 


Adi 




Eve 


Eve 


4. Apue 


Apue 


AbUHi 

(Abuni) 


Uva (Uvae) 


Uve 


^. Navlunu Naylunt« 


Narune 


ViU 


Idi 


<. Eja-m« 


Eja-m^ 


Teiamua 
: (Teiamera) \ 


Polem 
[Welegi] 


Pulegi 
(Puveage) 


7. Eli-nw 


Olime 


Eduma 


Polelu 


Polelu 


8. Etu-mtf 


Otu-me 


Ebuema 


Polee 


Pole 


9. Epue-m« Opue-m^ 


Napu 




Polouva 


Polohue 










[Polove 




10. Napnu 


Navlu 


Ekatoa 


Niigolu 
(Nukolu) 


r Nokolu { 
: Hokolu f 


4. YAIOKOLO 5. TANKMA 


6. TANBANU 


6a. vanikolo 


6b. vanixolo 


1. Tilu 


Kero 


lune 




Riro, Tasi 


Keru 


2. Taru 


Lalu (Lahu) 


Tilu 




Lai, Rua 


TaIu 


3. Telu 


Raru 


Teve 




Raru, Torn 


Rava 


4. Tava 


Rava (lava) 


Teva 




Rava, Fa 


LeU 


-6. TeU 


Teri 


TiU 




Sell, Rima 


Roo 


«. Tawo 


Ro 


Tuo 




Ro, Ono 


Roembi 


7. Tembi 


Rumbi 


Timb] 




Ruambe,Vitu 


Embidua 


8. Tawa 


( Lembidua \ 
\ Laubidua 


Tua 




Imbitua,Varu 


Duariudi 


9. Tanru 


Tuarendi 


Tindi 




Tauiine, Siwa 


— 


10. Kaluga { ^^1 } 


Tewaulu 


Edo«olo, 
Nofuru or 


Longolo 










JVafulu 




7. VINIKOLO 


7a. TUPUl. FBNUA OAL1.I0 8. UTUFUl. 9. WAMGO 


10. TBOXTA 




7b. 


tupua] 








1. TiLiogo 


Tuo Jika 




Sika 


Tai 


VuJi^ 


2. Ta Ru 


Buiu, (Biun) lu 




lu 


ERua 


Vu Ruif 


3. Te Lu 


Bogo Too 




Thoo 


E'Oru 


VuTd 


4. Ta Va 


Mabeo Jiva 




Sivia 


EHa'i 


VuVat 


6. Te Li 


Kaveri Jini { 


[Jun) 


Sin 


Rima 


T<fva lAme 


6. Ta Wo 


Elaveri-Juo Juo (Jaeo) 


Suo 


Ono 


L^Ji^ 


7. TeMbi 


Vio Timbi 


Timbi 


Bi'u 


Lev<?Ru^ 


8. Ta Pwa 


Viro Ta 




Ta 


Warn 


laeyeTel 


9. Ta Dru 


Reve(Rove) Tujo 




Toju 


Siwa 


Li Vat 


10. Gau 












Uluga 


A«aru iVavi 




.Vavi 


Ta«ahuru 


Hefwwul 




(Huaoi) 









Digitized by 



Google 



PofyauioH Linguitiie*. 



177 



11. B0TU1U. 


lU. XAia 

(XATBXA) 


12. pajDn 


13. TIKOFIl. 14. TONGA 


15. ILiOSX 


1. Ta 


Tai 


Tai 


Tasi, Taaa Taha 


Tahi 


2. Rna 


Lua 


Lua 


Boa Ua 


Rua 


S. Fola 


Tolu 


Tolu 


Torn Tolu 


Tora 


4. Heke 


Fa 


Fa 


Fa Fa 


Wha 


5. Lium 


Luna 


Lima 


Rima, Lima Kima 


Rima 


6. Ono 


Ono 


Ono 


Ono Ono 


Ono 


7. Hifu 


Fitn 


Fitu 


Fitu Fitu 


Whitu 


8. Walu 


Paru 


Valu 


Vara, Wara Valu 


Wara 


9. Siau 


Iva 


Eva 


Siva Hiva 


Iwa 


0. Sawhulu 


Kadoa 


Kado 


Fuaiiafuru Honofulu 


Tekau, 
iVahuru 



These Yooabularies are written in a uniform alphabet, a, e, i, o,. 
u as in Italian, a as English a in **bat," e as French e in *'le," o as- 
French eu in " leur," m as Frencli u in "lune," n as English ng in 
** sing," g as English ng in ** finger," j as English j in ** jail." 

In transcribing the Frencli vocabulary I have written w or u for 
ou, n for nh, j for dj and tch, and n for gn. 

These vocabularies are derived from the following sources : — 

1 and 7. Eev. C. E. Fox. Vocabularies of Santa Cruz and 
Vanikolo. 1908. 

2, 3 (in square brackets) and 9. Rev. R. H. Codrington, D.D^ 
Melanesian Languages. 1885. Words in brackets in No. 2 are front 
the Santa Cruz Prayer Book. 

3. MS. Rev. H. N. Drummond. In brackets from a M8^ 
written by J. Saii for Jtev. C. E. Fox. 

2a, 3a, 4, 5, 6, 7a, 7b, Ua. Dumont D'Urville. Voyage de^ 
TAstrolabe. Philologie. Tome 11. 1830-34. Words in brackets 
are from P. A. Lesson in Revue d' Anthropologic. 1876. 

6a. Rev. J. Inglis. Report of a Mission Tour in the New^ 
Hebrides. Journal Etlinological Society. 1864. 

6b. Journal of Commodore Goodenough. 1876, 

8. MS. Rev. B. Teilo. 

10. MS. written by A. Towia for Rev. C. E. Fox. 

11. MS. W. L. Allardyce, C.M.G. 

12. MSS. Rev. J. W. Blencowe, Rev. H. N. Drummond an* 
N. Vane. 

13. Rev. W. J. Durrad. Tikopia Vocabulary. Journal Polynesian* 
Society, XXII. Words in brackets from W. H. R. Rivers. History 
of Melanesian Society. 

14. Rev. W. Shirley Baker. English and Tongan Vocabulary ». 
1897. 

15. Rev. H. W. Williams. Dictionary of the Maori Language.. 
1917. 



Digitized by 



Google 



GII,BERT'S ACCOUNT OF EASTER 
ISLAND. 



By H. D. Skinner. 



THE present writer was statioiiied in England during 1916 and part 
of 1917, and spent part of his spare time in collecting inforuiatifm 
relating to the ethnographic material from the Pacific brought to 
England by the members of the three expeditions led by Captain 
Cook. In the Bntish Museum and at the Records Office, Chancery 
Lane, London, he found the logs and journals of a number of officers 
who sailed on one or another of the voyages, documents which, in so 
far as they relate to New Zealand, have already been published by the 
late Dr. BK)bert McNab.* The far larger part of them, however, which 
relates to Polynesia, Melanesia, North America, and Alaska, but 
especially to the first-named region remains unpublished. It seemed 
to the wiiter that if funds were available the publication of these 
journals ought to be undertaken by the Polynesian Society in the form 
of a Memoir or Memoirs unifomi with those they have already 
published. Publication in some form is called for, first by their historic 
interest, and secondly by the amount of new ethnographic matter 
they contain. 

The following extract and sketch were copied from the journal of 
Jos. Gilbert, master of the '^ Resolution," and are of unusual interest 
in that they relate to Easter Island, which on account of its gi'eat 
statues has attracted keener public interest than any of the smaller 
islands of the Pacific. At the Royal Naval College at Greenwich are 
the originals of a number of portraits by Hodges, including two of 
Easter Islanders, engravings of which were published in the first 
■edition of the second voyage. 

The relevant part of Gilbert's joui'nal is as follows : — 

^* Wednesday, 16th March, 1774. 

" The natives are of a niiddle stature, well made, of a copper colour 
complexion, with a brisk and lively countenance, very active and 
audacious. Some wearing a light garment over their shoulders, others 
quite naked, women few have been seen, those kind and obliging 
wearing a thin light covering round their waist hanging carelessly 
down to their knees. Both men and women paint with a bad kind of 
Vermillion; besides heavy, exceedingly curious figures and lines marked 
upon their .faces, legs, arms and different parts of the body by a 
black liquid punctuated thro' the skin. 

The islanders throughout the Pacific Ocean have this custom, and 
in general their rank in life distinguished by different characters. 

* '♦ Historical Records of New Zealand/' Vols. I. and II. 



Digitized by 



Google 



Gilberts Account of Easter Island. 179 

At Otabeiti the young women were not pemiitted to marry till this 
ceremony is pei-fonned, which is from the loins round their hips, and is 
continued to tlie lower part of the thigli behind ; curious an4 
exceeding handsome, nay, so becoming in those people that to an 
European they would appear naked without it. 

The land is extremely poor, the iiills full of atones, of an hungiy 
dry soil incapable' of cultivation. Some plantations of potatoes and 
plantains to be met with in the valleys, which are the principal 
productions. No trees of any size sufficient to make^ore than the 
helve of an axe, nor any kind of vegetable useful for refreshment. 
Water exceeding scarce and indifferent throughout the island. The 
inhabitants do not I believe exceed a thousand men. During our stay 
here we got five tuns of bad water, some plantains, potatoes, and a few 
fowls. The wind changing to the westward the captain thought it 
necessary to put to sea, otherwise we might have purchased a sufficient 
quantity of roots to have served the ship's company four or ^ve weeks. 
Cocoanut shells andOtaheiti cloth are the best commodities for ti'aid ; old 
hats, rags and bottles are not bad things. Fish seem scarce, we caught 
none ; no quadruped animals except rats. Providence has even denied 
tiieni the pleasing service and companionship of the faithful dog. What 
I liave seen of their habitations are of an ovil form about 10 or 12 ft. 
in length, in height 4 or 5 ft., made of rushes, reeds and plantain 
leaves, strengthened with a few sticks fastened in a platform of stone 
on which they are built. At one side is added a small porch through 
which they Creep into the hut ; no furniture observed within, only a 
little straw, on which they compose themselves at night. 

Their canoes are as indifferent as their huts. From 20 to 25 feet in 
length and a foot and a half in breadth, quite open, made of thin 
boards, badly put together. We have some reason to believe these 
islanders to have once been in a more flourishing state. There remain 
some vestiges of exceeding curious workmanship, which to appearance 
must have been executed some centuries back, and seem impossible for 
those people in their present helpless situation to equal. These are 
statues of a monstrous size erected along the seashore. Time having 
brought several to the ground ; some 27 feet in length, breadth in 
proportion, one solid mass of stone, the weiglit having been calculated 
to exceed 20 tons. It required great ingenuity to have erected these 
prodigious images without the knowledge of machine power ; the 
wedge, lever, pulley and screw are not in their possession. Neither does 
the island visibly produce materials to make them. And what is more 
extraordinary these figures are capped with a stone larger than the 
circumference of the body altogether, making a very formidable 
appearance. Before these images a square out of stones artfully fitted 
into each other, raised 6 or 7 feet in height in a direct line from 30 to 
40 feet betwixt the facings." 



Digitized by 



Google 



NOTES AND QUERIES, 



^ [290] The Wanaka District. 

Mr. Richard Norman, of Oamani, formerly of Wanaka, tells me that when 
the first white settlers visited the Makarora Valley, they saw there a number of 
birch trees, the bark of which had been scored down one side by the Maoris, 
the intention being to cause the heart-timber to rot while the outer timber 
continued growing, thus half forming a canoe. On the bank of the Clutha there 
were conspicuously placed a number of white quartz boulders, each about the size 
of a loaf. The Maori shearers explained these as a message indicating that a 
party of Maoris had passed, going to Hawea, and would return. Of a large 
lagoon on the east side of Lake Hawea the following story was told : A tohuka 
was fishing and caught a huge eel. In carrying it home he became tired and cut 
it in half, taking one half home. On returning for the other half he found it had 
turned into the lagoon, and that its blood was represented by the stream which 
flows into Lake Hawea. 

These fragmentary notes are worth recording as we know almost nothing of 
the Wanaka district in pre-European times. 

H. D. Seinnbb. 



Digitized by 



Google 



PROCEEDINGS. 
POLYNESIAN SOCIFIY. 



A Mbstiko of the Council took place at the Library, Hempton Room, on the 3rd 
September, when there were present : The President in the chair, and Messrs. J. 
B. Roy, P. J. H. White, a. H. Bullard, M. Fraser, W. W. Smith, W. L. 
Newman. 

After the minutes of the previous meeting had been read, the following new 
members were elected : — 

John Baillie, Director, New Plymouth Museum. 

G^rge N. Curtis, Stratford. 

H. M. Good, Stratford. 

Alfred Snowball, The Manse, Ormond, Gisbome. 

The Invercargfill Public Library, LiTen»ugill. 

Onehunga Carnegie Public Library, Onehanga. 

G. T. Kronfeld, P.O. Box 405, AucWaiid. 

Wm. McKay, F.R.C.S.E., 46 Guinness Street, Greymouth. 

James McKay, L.S., P.O. Box 65, Greymouth. 

The foUowing papers were received : — 

History and Traditions of Rarotong^. Parts VIII., EX., X., XI., XIII., 

XIV. 
Native Names of Birds, Mangaia Islands. F. W. Christian. 
An Ancient Maori Pare. 

Traditions and Legends of Murihiku. H. Beattie^ 
The Rua-kopiha. G. Graham. 
It was agpreed to exchange with the Museu Paulista, Coixa g. San Pedro, 
Brazil. 

The death of one of our original members, the Hon. Sir John Dennison, was 
reported ; and the resignation of Mr. Drummond was received. 
A list of books, etc., received in exchange was read. 



Digitized by 



Google 



Digitized by 



Google 



HISTORY AND TRADITIONS OF RAROTONGA. 



By Te Ariki-tara-ake. 



PAET VII. 

Translated by 8. Percy Smith. 

[In Part VI. of this series the lives, emigrations and voyages of the 
ancestor of the great Karotongan chief, Tangiia-nui were recorded so 
far as the information to hand allowed. That part brought the history 
down to the thirteenth century, when Taugiia's immediate foi'beara 
had settled in Tahiti. This, Part VII., will describe the life and 
adventures of Taiigiia-nui from his birth to the time — probably 
in his middle age — when he settled down in Rarotonga. Perhaps the 
distinguishing feature of this portion of his life is the number of 
voyages he made in his celebrated vessel, '*Takipu*' (afterwards 
renamed **Taki-tumu," though tiiis latter name must not be confounded 
with another vessel of the same name that came to New Zealand in 
the middle of the fourteenth centur}'). 

Tangiia's voyages illustrate in a striking manner the ability of 
these South Sea vikings to make very extensive voyages, by which, 
as their own historians say, they became able navigators, and that at 
a period when no other nation in the world had accomplished 
anything more than what may be termed coastal voyages. Our 
historian treats of these voyages in a manner to indicate that the 
people thought nothing of starting o£E on expeditions covering some 
thousands of miles. No details of the voyages are given ; even the 
class of vessel employed is not stated, any more than that they were 
pdH, which we know to mean a large, built-up canoe, generally a double 
one with a platform between them ; though from what is stated in 
paragraph 307 the inference is that Tangiia made most of his voyages 
in an outrigger canoe. The following brief abstract of Tangiia'& 
voyages will show the extent of his travels : — 



Digitized by 



Google 



184 JOURNAL OF THE POLYHESIAH SOCIETY. 







Approximate 


ragraph. 


Extent of Voyage. 


Nautical Miles. 


282-3 


Taliiti to Mauke 


480 


284 


Mauke to Taliiti 


480 


285 


Tahiti to Savaii, Samoa 


1320 


285 


Back to Tahiti 


1320 


294 


Tahiti to Samoa and Avaiki (to Savaii) 1380 


300 


Avaiki and hack to Ilea Island. 


? 


300 


Uea to Upolu 


270 


300 


Upolu to Uea and back 


540 


301 


Upolu to Fiji (? Lau Islands) 


480 


302 


Fiji to Easter Island 


4200 


303 


Easter Island to Moorea Island 


2400 


304 


Moorea to Huahitie and Porapora 


150 


310 


Porapora to Fiji 


1680 


313 


Fiji to Paumotu Islands 


2400 


316 


Paumotu to Taha*a Island 


720 


317 


Taha*a to Rarotonga 


540 




18,360 



To this considerable mileage — measured in an air-line, though 
there is no doubt the course taken against the trade winds would 
involve much longer distances — we may fairly add the distance run to 
the south of Itarotonga when they missed that island and found the 
flea- water very cold — probably another 800 or 1000 miles at least. 
Again there was the voyage to Avaiki, and as the position of this 
particular Avaiki is doubtful, we cannot estimate a mileage for this, 
probably the longest voyage of all. The name is given both as 
Avaiki and Avaiki-te-varinga (the ancient Avaiki), and it is clear that 
this was not Avaiki-Atia, the original Fatherland of the people (more 
generally called Atia). In the book ** Hawaiki," it was attempted to 
identify this Avaiki with Java or Sumatra, but certain considerations 
lead me to think it may have been some other of the Indonesian 
islands, perhaps the Celebes or Gilolo, or even Borneo. But this is 
not the place to discuss that question. 

I fear I have not always understood some of the detail in the 
Sage's story.] 

TANGIIA's birth and disputes at TAHITI. 

WHEN Tangiia was born his grandfather, Ka*u-kura, gave him 
the name of Rangi, and when his uncle, Pou-vananga-roa, 
learnt that his sister, Ka'u-ngaki, had given birth to a son, he 
adopted the boy for his own, and renamed him Tangiia-nui. The uncle 
also adopted the son of his other sister, Maonga, named Tu-tapu, as 



Digitized by 



Google 



Hhtory and Traditions of Rotoionga. 185 

a second child. They tlnis became tlie adopted children of Poii- 
vananga-roa. [We learn from other sources that Tangiia's original 
name was Uenga.] 

279. Tiiis is the line of descent to Tu-tapu (on his fatlier's side) : 
Nga-upoko-turua-o-Taie and Tai-ranu were tlie sons of Onokura by 
his wife Te Ata-nna.^ The first named son had Nga-maru, who liad 
Kotuku-tea, who had Pou-tea, wlio married Maonga, the sister of 
Pou-vananga-roa, and tlieir son was Tu-tapu, who was thus an elder 
brother (really first cousin) of Tangiia, while Mnono (eldest son of 
Pou) became the ariki, and his brother Ken a priest. 

280. This is the history of these children of Pou-vananga-roa:^ 
He appointed his children to their various functions, Maono to be 
arikiy or head chief, Tu-tapu was also to be an ariki^ while Tangiia 
was to be a tavana, or minor chief. Hence grew up dissensions because 
tlie latter objected that he was not also made an ariki. Tu-tapu was 
an ariki of Iva (the Marquesas), while Maono was an ariki of Tahiti. 
Tangiia became much enraged, and seized the insignia of rank (or 
usurped the position) of the ariki at the marae at Avarua, and 
performed the functions, driving Maono away to the mountains. The 
village now became his and part of the kuru (probably the first fruits 
of the breadfruit, as a right of the chiefs). Maono fled through this 
persecution to the hills, and then the evils commenced. When Tangiia 
had seized the power and the lands, his younger bretliren awaited (in 
vain) the distribution of part of the lands and some of the power for 
themselves to be given to them by Tangiia ; and this remained a 
source of evil between them. These are the names of those people : 
Nuku, Ai, Ika, Mere, Uki and Maraka. 

281. The second cause of trouble between Tangiia and Tu-tapu, 
was a stream, Vai-iria, a sacred stream of Tangiia's. Trouble broke 
out over that stream in which Tu-tapu bathed, and fighting followed 
at Patu-te-ere-tiki and Umu-toto-tatatou, where Tangiia's "teeth 
were broken " and the plumes of Te Tua-ki-taaro were smashed- 
This is the meaning of Te Tua-ki-taaro, it means (the people of) 
Avaiki, while Te Tua-ki-taa-poto means the (people of) Iva and Tahiti. 

1 . Onokura is a very celebrated ancestor of these people, whose adventures fill 
forty-seven closely written pages of the documents we are translating, and these 
we hope to translate later on. 

2. There is a very strange story embodied in the MSS. from which these 
traditions are taken, relating to one Kiri-paru, a descendant of the above 
Pou-vanangaroa-ki-Iva (which is his full name, and may be translated as Pou-of- 
the-lengthy-recitation-at-the-Marquesas Islands). During one of the numerous 
intertribal disputes on the question of ritual at the maraes^ Kiri-paru, who appears 
to have been a member of the people living at Rangi-atea on ^ the N.E. coast of 
Rarotonga, in order to accomplish his object stayed the course of the sun in its 
orbit ! This same Kiri-paru seems also to have been endowed by great powers of 
sorcery, as we shall see later on when we come to deal with his family. 



Digitized by 



Google 



186 JOURNAL OF THE POLYNESIAN SOCIETY. 

282. The third cause of trouhle was ahout the two fisli, the 
raratea (the sliark tribute) and the onu (the turtle tribute). Tu-tapu 
insisted that when a shark was killed the head was his tribute, while 
the tail-end was for Tangiia, and that the turtle sliould be divided 
one part for him, one for Tangiia. But Tangiia would not consent to 
this; and further trouble arose between them, leading to their 
separation. Tu-tapu then returned to Iva (Marquesas) while Tangiia 
started on a voyage to Mauke Island (of the Cook Group), to visit the 
•daughters of Te Tata-uru-ariki, and his wife Te Puaranga-uta, 
which daughters were named Pua-tara and Moe-tuma. 

TANGIIA VISITS MAUKE ISLAND. 

283. When Tangiia reached Mauke Island he landed in the 
district of Utaki, wliere is the hoidu (or sacred place of religious 
ceremonies, meetings, etc., etc.) named Eangi-manuka. He went to 
look for the girls, and found them by the path beating out bark (for 
<5loth). He stealthly drew near to them, through the bushes, and as 
he approached he felt inspired and composed the following love 
song : — 

[Which song I do not feel competent to translate — see the 
original.] 

WAE AT TAHITI BEl'WEEN TANGIIA AND TU-TAPU. 

284. After these proceedings he returned to Tahiti.^ His sister 
asked him about the hiru (breadfruit tribute). He replied, " I did 
not pluck it." She had thought that Tangiia had it, and was not 
aware that Tu-tapu had taken it away with him to Iva. In conse- 
quence of this a quarrel arose between the brother and sister in front 
of the marae, which grew in violence between him and Maa, the 
sister's husband [other sources give the husband's name as Uki- 
mara].* Tangiia snatched part of the turtle from the hand of Maa, 
And this caused ill-feeling on the part of the latter. So the sister 
(named Raka-nui) took the canoe named ** Kai-oi" and departed for 
Huahine Island, accompanied by Maa. 

285. Tangiia now dwelt at Tahiti [from what follows, it must 
have been for some years], and then he fitted out the canoe named 
**Tuna-moe-vai," which name had been given by his grandfather, 
Ka*u-kura [who had also been a noted voyager see ** Journal 
Polynesian Society," Vol. lY., p. 102, where, however, Mr. Stair calls 

3. The narrative does not say if Tang^a was successful in his love making to 
these two ladies, but we shall learn further on that he had children by them. 
Their names were Moetuma and Puatara. 

♦ From the same source we learn that these two were the parents of Tarionge, 
Au ancestor of the Tarauaki tribe of New Zealand. 



Digitized by 



Google 



Digitized by 



Google 



Pa •? 




^ 



< 













3 



o 



Digitized by 



Google 



Hisioiy and Tradiiions of Raroionga. 187 

i;he voyages tkere described " Sanioan Voyages," though in reality 
they were made by these people we are dealing with], while his 
fiister named it ** Te Tika-o-te-tuaine.** On completion Tangiia 
renamed it ** Taki-pu," and then he visited Avaiki (Samoa) and many 
other islands. On Ids return to Tahiti he dispatched Tino-rere to fetch 
his children from Mauke Island. 

286. On the latter's return from Mauke, Tu-tapu and his host 
liad arrived from Iva (Marquesas). The object of Tu-tapu's expedi- 
tion was to make war and to secure tlie celebrated weai>on named 
^*Te Amio-eniia" * from Pou-vananga-roa (his adopted father), and 
the rara-roroa [emblematical for a man ? for sacrifice]. But Tangiia 
emphatically refused to concede these things, the weapon and tlie two 
**' fish " ; Tu-tapu holding that they were liis [by righ of seniority], 
but Tangiia was obdurate. Tu-tapu then abandoned these things and 
-demanded the rara-kuru [the breadfruit branch — some right of a 
<;hief, not known to the translator], and this Tangiia conceded in 
order to pacify Tu-tapu, but it had no effect on the latter. 

287. And now war commenced ; Tangiia assembled his clans the 
Kaki-poto, the Atu-taka-pofco, the Kopa, the Tavake-moe-rangi, the 
Tavake-oraurau, the Neke, the Ataata-apua and the Tata-vere-nioe- 
papa.* There was now a separation of the people, Tu-tapu and his 
army retiring to Tautira [a beautiful place on the north sliore of the 
■eastern peninsula of Tahiti], while Tangiia retired to Puna-auia* [a 
river and district at the extreme west part of Tahiti] . 

288. Tangiia tlien composed the following lament for his land; — 
[See tlie original , I cannot translate it, though the sense can be made 
out. Tangiia evidently forsaw disaster to his party in the war about 
to commence, and heuce the lament.] 

289. The war now commenced and great was the destruction ; 
one side of Tahiti was full of Iva people. Tangiia retreated, followed 
up by the Iva forces. So Tangiia gathered all his people together, 
a-nd prepared their vessel, launched her, and placed on board all their 
property, including the (emblems of tlie) gods Tonga-iti, Rongo-ma- 

4. For the orig^ of this weapon see Part VI. hereof. It is clear that for part 
of his life Tu-tapu lived at Afareaitu on the east side of Moorea Island, for in 1897 
I was shown the stone foundation of his house. 

5. It seems prohable from many circumstances that these clans, or some of 
them, were Melanesians, brought from the west as crews of the vessels. See the 
description of some of the Ra'iatea people of this period in our ** Memoirs,** Vol. 
IV. 

♦The accompanying sketch shows Te Fana-i-Ahurai district at the ex- 
treme west end of Tahiti, and Puna-auia, Tangiia*s home there, and the hills that 
were set fire to. There is a level flat along the sea, varying from a quarter to a 
half mile wide. The forts were used by the French in their war with the Tahitians 
in the forties of last century. 



Digitized by 



Google 



188 JOURNAL OF THE POLYNESIAN SOCIETY. 

Tane, Rua-nuku, Tu, and Tangaroa. Two of the gods had been 
taken by Tu*tapu, viz., Rongo-raa-Uenga atid Maru-mamao. The^ 
chief's seat named ** Kai-ati-unga " was also secured. After all these 
things had been placed on board, they returned and continued the^ 
fighting, and then, on the mountains, fell liis cliildren Pou-te-anuanua. 
and Motoro. 

290. This was tlie cause througli which his children fell. During 
the course of the battle instructions were given to his cousins Uki 
and Maraka, as follows: **At Vai-tietie, my cousins! place your 
company." To Ika and Mere lie said, ** Station yourselves, your 
young relatives and your companies at Vai-pa-ekeeke." To Nuku 
and Ai, ** At Vai-pikira you and your junioi's and companies station 
yourselves." These streams were the fighting places, where the 
assault took place, and here the ridges were set tire to [? by Tu-tapu*s 
people] and caused the death by burning of Pou-te-anuanua. All 
Tangiia's fighting streams were taken, and he (and his people) were 
driven into the sea. Tangiia reached the sea, and waded over the 
shelving reef, got on board the canoe. From there he beheld all 
inland covered with lire and smoke, right Tip to the upper mountain 
spurs. 

291. Now, the goddess Taa-kura, looking down on the raging 
fire, saw Motoro [Tangiia's son] enclosed by the fire. She said unto 
the god Tangaroa, ** Alas ! this ariki ! He will be burnt by the fire ! " 
Tangaroa replied, **What is to be done? You are a god; he is a 
man." ** Eia ! I will haste and fetch my husband! " said Taa-kura. 
Then Tangaroa urged her, saying, ** Haste thee to Retu^ and ask 
him to give you a pakoko-ivi (a thick cloak worn on the shoulders) to 
cover him with." She also asked for a gust of favourable wind in 
which to hide herself in her descent, and then the south-east wind 
was sent, into which the goddess entered and covered Motoro with the 
cloak ; then seizing him she took him off to A*ua*u [the island now 
called Mangaia, where Motoro was afterwards deified. For reference 
to him see *' Rarotonga Records," p. 28]. It was the Muu and the 
Pepe (the latter is the butterfly) who carried him there. And hence 
(? were these names) given to Tinomana. [But other stories differ 
from this. Thej' say that Motoro remained iii Tahiti for many years, 
until he was brought from there to Rarotonga after the war, described 
in Part IX., by Keu the priest, after calling at Mangaia Island^ 
where another son of Tangiia's, named Te Rei, had arrived from 
Mauke Island. Motoro became the ancestor of the Tinomana family 
which still lives at Arorangi on the west side of Rarotonga.] 



6. It is not known who this personage is — probably a god. 



Digitized by 



Google 



History and Traditions of Rarotonga. 189 

292. After Taiigiia had gazed for some time at the (burning) 
land, he lamented it as follows : — 

Great is my love to my land 

Tahiti that I am leaving 

And geat is my love to my koutu 

At Pureora, that I am leaving, 

Great is my love for my drinking springs, 

Yai-knra-a-mata and Marama-ata-kai 

That I am leaving, 

My bathing places, Vai-iria and Vai-te-pai 

That I am leaving, 

My own district, Piina-auia and Papeete 

That I am leaving. 

For my moontaiiis, Tikura-mammara 

And Aorangfi, that I am leaving 

And for my dear children 

Pou-te-anuanua and Motoro now dead 

Alas ! Alas ! O my children ! 

My children O! Alas! 

O Pou-te-anuanoa I Alas ! 

O Motoro O ! O Motoro ! 

293. After this, Tuiti and Niikua-ki-roto were sent to the marae 
to fetch a wreath and some red berries, and certain white tapa [used 
in the marae\ which things were gods of theirs. But they did not 
-go to fetch those things, but rather to steal the god Ronga-ma- 
TJenga to be taken on board the vessel. It was in consequence of this 
theft that Tu-tapu so persistently pursued Tangiia [as we shall see]. 

TANGIIA STARTS ON HIS VOYAGES. 

294. The vessel now departed to the west to all the islands there, 
to Avaiki (Savai*i, Samoa), Tangiia still lamenting the loss of his 
<5hildren and his home. His gi-ief for his children was something 
-extraordinary. Pai [Tamarua-pai, see Part VI. hereof] joined in this 
expedition, for Tangiia had taken him as his chief warrior (and 
navigator). It was his function to navigate the vessel to Avaiki, and 
"Tangiia gave him the name of Tei-vao-ariki.^ 

When the vessel reached Avaiki [not 8avai*i, but in Indonesia 
probably], the trumpets and the drums were heard, and Tangiia 
asked, " What are these trumpets and drums sounding for ? " Pai 
Teplied, ** They are trumpets and drums denoting the appointment of 
-a chief, and the takurua feast, and tlie accompanying ceremonies and 
-dances ; a gathering of the many gods to the feast, with their priests. 
It is that which we hear." 



7. It would seem from this that Tamarua-pai had already visited this distant 
Avaiki, which is certainly not 8avai*i of Samoa. 



Digitized by 



Google 



190 JOUBHAL OF THE POLYKESIAH SOCIETY. 

THEY INTERVIEW THE GODS. 

295. They then went ashore, and there Te Mata-iri-o-puna and 
Pai were sent to (the gods) Tu-te-rangi-marama, Rua-i-te-kari, Tii,. 
Tangaroa, Rua-te-atonga, Taakura, Tupua-nui, Rangatira, Tii-vaero- 
kura, Kura-akaipo, Aronga, Tu-Avaiki, Tu-tavake, Rongo, Ueiiga, 
Maru-mamao, Kau-kura, Tonga-iti, Nge-mua and Mumai-io. Their 
gatliering was the celebration of the takurua [presentation of first 
fruits] at this time, and it was just then that Tangiia arrived at 
Avaiki from Tahiti after the war with Tu-tapu, and death of his 
children. 

296. Tu-te-rangi-marama asked of him, ** Why have you come 
to Avaiki?" Tangiia replied, ** My coming is due to misfortune ;^ 
Tu-tapu and I have been at war, and my children Pou and Motoro 
are dead, whilst I had to flee from Tahiti with all my propeiiiy. Aa 
for this, I am at a loss what to do." Then all that gathering pro- 
ceeded to enquire of him, ** What was the cause of your war ? " He 
confessed to them, " It was the government, the ariki-shi'p ; the fish- 
tribute ; the turtle -tribute ; the shark-tribute and the human-tribute ^ 
besides the weapon "Te Amio-enua" — these were the fruits [causes] 
of our war." 

Tu-te-rangi-marama then spoke, saying, **It was lucky you were 
tapuy and were not struck " [? wounded or killed]. Then they 
consulted among themselves, saying, ** Let us give unto him great 
powers, and set aside some particular land for him to dwell in.'^ 
Then Tonga-iti said to Tangiia, ** There is a home for us two, named 
Tumu-te-varovaro [Rarotoiiga, an ancient name] ; go thou to that 
land, and there live till thou diest." And then they gave unto him 
great mdna [power, wisdom, prestige] which is the ira ^ of the gods, 
the meaning of which is tliat he would always conquer in war — he 
would always win. 

297. After this he begged of the gods that they would come with 
his expedition.^ These are those companies and their armies and 
their followers: Tangaroa and his company; Tu-te-rangi-marama and 
his, and the same for the following gods — Tu-tavake, Rua-i-te-kari^ 
Kau-kura, Mumai-io (who is the same as Tou-tika, and his descen- 
dants, seven in number, three sons and four (laughters). Rua-te-atonga, 
Ari, Tupua-nui, Taa-kura, Kura-akaipo, Tonga-iti, Rangi-puta-rua, 
Ai-marie, Maru-mamao. And now for the first time did Tangiia 
possess all these properties. 



8. /r«, in Maori means the ** spark of life,** but here it probably means 
upematural powers. 

9. Apparently this means that some figures of the gods were to be given to 
him. 



Digitized by 



Google 



Mistory ami Tratfithns 9f Rutotongu. 191 

[The wliole of paragrnplis 296 and 297 »ire difficult to understand. 
Tlie Sage writes as if Tangiia had actually interviewed tlie gods- 
themselyes, whereas it probably means that he visited tlie priests of 
those gods in this distant Avaiki, which is not Savaii, but very likely 
Java (Hava, Ava, Ava-iki), or some other part of Indonesia whero 
there were still living some of their own race, left behind when the 
great migration took place — see introduction to this part. These 
* gods ' he brought away were probably figures of them such as are 
used by all Polynesians.] 

298. Tangiia then begged (of the go<U) for all things connected 
with the takuruaSj which application was consented to. These are those 
yarious things : — 

Te eivUf five kinds The pen au eiva 

A kaara The akaingo 

A puapua-aki\ or tioi The pat* 

A tutmimaa A kapo-rakau 

An eva-tipa A puputa 

A Taumua ponpou A pare-eva-tipa 

[These are various kinds of posture-dances, drums, etc., some of 
whicli are described in Par. No. 369.] 

299. When these things came into his possession Tangiia ordered 
Taote and Mata-iri-o-puna to take charge of the trumpet and the 
drum. And to Tavake-oraurau he assigned the charge of the things 
pertaining to the takuruas^ the ara o nga atua (?), and a drum to Avaiki. 
Anga-takurua took Tangiia's place in the direction of the takurua 
(ceremonies) and Te Avaro, from Rangiraro Island, took charge of 
other trumpets on board the vessel. Tangiia had now secured all 
there was to be got at Avaiki, and hence is the saying, ^^ Ko te tere a 
te tumu " (the expedition that cleared out all down to the bottom — 
Free translation). 

THE EXPEDITION RETURNS SOUTH. 

300. The vessel now returned, and on its way called in at Ilea [or 
Wallis island, north-oast of the Fiji Group; another name for this 
island is Varehao]. Here they beat the drums and sounded their 
trumpets. On departing, Te Avaro left behind one of the trumpets. 
The vessel came on from there and reached Kuporu (Upolu, Samoa)^ 
where tliey again beat their drums and sounded the trumpets. 
Behold ! there was no trumpet because Te Avaro had left it at Ilea. 
They returned at once to Uea to fetch it, and on securing it Katu 
joined them. Back again to Kuporu came the vessel and there beat 
the drums and sounded the trumpets, whilst the following song was 
composed. [Though I cannot translate the song in full, the meaning 
of parts is quite plain. It refers to the acquisition of the drums, etc., 
and states that one was made from the tree cut down by Ui-te- 
rangiora, etc., etc.] 



Digitized by 



Google 



192 JOURNAL OF THE POLY M ESI AM SOCIETY. 

301. The expedition sailed for Iti [Fiji — probably the Lau 
Islands — where no doutt some of tlieir compatriots were still living] 
and here they met Iro f Whiro, a well-known ancestor of Rarotongans, 
TahitianSy and Maoris], Taugiia asked him, ** Where is thy child 
whom I want as an ariki (for my people)? My children are 
•dead." Iro replied, " O ! He is away at distant Eapa Island ; I liaye 
settled him there." Tangiia then said to Iro, '* I shall go there and 
take him as an ariki for my people." To this Iro consented, saying, 
** Go then, and take him." 

EASTER ISLAND. 

302. The vessel now went right away to Rapa-nui [Easter 
Island, which name was given to it to distinguish it from Rapa-iti, 
south-east of Rai'otonga, and from the latter island it is said some of 
the Easter Islanders sprung. But the native name of the island is 
*re Pito-te-henua, also Vaihu, Mataki-te-rangi and Kairangi]. On 
i.heir arrival they found this boy, by name Taputapu-atea, who was 
on the reef diving for shell-fish, to whom Tangiia addressed himself, 
«ayiiig, **Who art thou?" He replied, "I am Taputapu-atea!" 
Tlien said Tangiia, ** It is well, let us go, for I have come for you, 
iind will make you an ariki" And so that child was taken on board 
the vessel, which tlien departed. 

TANGIIA AND HIS SISTER MAKE PEACE. 

303. The vessel then proceeded to Moorea Island [just fifteen 
miles to the west of Tahiti, the most beautiful island in the Pacific], 
but they found that Iro had not arrived, so a message was left, ''If 
Iro should come here, tell him that Tangiia has taken his son to 
make of him an arikt,*^ 

The vessel now sailed for Uaine [Huahine, about eighty miles 
W.N.W. of Tahiti], and on arrival Maa [Tangiia's brother-in-law'**] 
•came on board to see Tangiia, and said to him, ** You had better 
return, for thy sister Raka-nui has prepared and dried the firewood to 
<50ok you with." Tangiia replied, ** I shall not return, I am the priest 
TJnutea, and my assistant is Marotai, I insist on going ashore." 
When he had entered the channel (in the reef), Raka-nui and her 
•daughters came and began to abuse him [it will be seen in paragraph 
284 that they had parted in anger when they last met] saying, ** 
thou great big headed eel! fall down and grovel in the water. 
Behold ! my brother ! You have been brought here to my place, 
in the great ofcean. That is the direful result, my brother ! of evil." 

10. According to the ^'Tiiatua no Tangiia-uui," p. 2, the husband of Raka- 
nui was Uki-maraka, probably another name for Maa, and their son wasTarionge, 
■an ancestor of the Taranaki Maoris. 



Digitized by 



Google 



Hittoij and Trm/itions of Raroionga. 19S 

Then those on board the vessel answered the abuse of those on sliore^ 
to the Papaka-tueraera, saying, ''It was thou who brought (the- 
trouble ?) to Hualiine and Ngangaua. Uki and Kakaraea dwelt with 
tlie woman Raka-nui, and gave birtli to Tarionge and Te Ariki-oa- 
ngata, that searched for some /^ani at Taliiti. This is I, Tangiia, a 
priest Unutea, whose assistant is Marotai. I have a very fine cano& 
now, so was the way to Huahine made eauy to my sister Haka-nui. 
Let me go ashore. Keu [their brotlier] is a^ore there." On tliis 
the sister cried over Tangiia and rubbed noses, and then they went to 
the village. 

- [The greater paH of par. 304 is very obscure, and probably I 
have missed the meaning of these emblematical sayings. It is 
interesting to note the name of Tarionge, who was one of the ancestors- 
of the Taranaki people of New Zealand, many of whom at this period 
vrere dwelling in the neighbouring island of Eangiatea (Ra*iatea).] 

305. When they arrived at Eaka-nui's house, a feast was. 
prepared for Tangiia and the crew of the vessel, and after hunger had 
been satisfied, Raka-nui questioned Tangiia as follows, ** What has 
caused your delay ? What have you been doing ? How is <mr home 
(in Tahiti)? Where is our rara-kurUy and our gods? Where is the 
weapon, * Te Amio-enua' ? and our children, and our father?" [In 
reference to the rara-kuru, or breadfruit tree branch, I was told, by 
the Rev. W. Wyatt Gill I think, that in the times of Tangiia the- 
population of Tahiti had become so dense that the product of bread- 
fruit trees was divided and owned by separate families — no one was 
allowed to own the whole of the fruit from one tree. And that a 
dispute between Tangiia and Tu-tapu over a rara-kuru was one of the 
causes of their subsequent war. It is probably this that is referred to 
above. It is to be remembered that the breadfruit is*iiot indigenous- 
to the Pacific, but was brought by the people from Indonesia, and 
lience perhaps in Tangiia's time it had not had time to become so 
plentiful as in these days.] Tangiia replied to his sister, ''I have been 
delayed by my hmg voyage over the ocean, to assuage my anger [at 
his defeat, see par. 290-291, etc.] and to mourn the loss of my 
children." The sister asked, ** Are your children dead?" **Both 
Pou-te-anuanua and Motoro are dead ; the land has been conquered ; 
our rara-kuru is gone ; as also one of our gods." " But where then is 
the weapon ? " ** It is with me ! " Said the sister, ** What then has 
been the real cause of your delay ? " Then Tangiia explained to her, 
** I was a long time at Avaiki-te-varinga." ** A ! have you really 
been to distant Avaiki ? What is the news from there ? " Tangiia 
replied, **This is the news that I have gathered at Avaiki ; there was 
a great taknrna (feast and ceremonies) going on when I arrived there,, 
accompanied by all kinds of games and amusements." 



Digitized by 



Google 



194 JOURMAL aF THE ROLYMESfM ^BCfETY. 

306. Then Haka-nui asked of Taiigiia, ** Describe those things 
which you brought back from there.'' Tangiia replied, '* These are 
the whole ol them ; we brought away everytliing ; there is notliing 
left." llaka-nui asked, '* Where is the eva-ttpa (ceremonial dance) ?" 
*' It is here." ** O ! let us see it ! " 

Then the sister tided to pursuade Tangiia to remain there at 
Huahine, saying, ** Le^ us both live liere. You on one side of the 
island; I on tlie other." But he replied, *'Not so! I cannot remain, 
I must go on to Tumu-te-varovaro [old, or spiiit-name of Rarotonga], 
which (the god) Tongo-iti has told me of. I am going on there to live 
tUl I die." << Wliat land is that? " asked the sister. Said Tangiia, 
** What land indeed! I have never seen it." 

Then Tangiia explained to his sister, ** With me is the son of Iro, 
whom I have brought along to make an ariki of." ** Why is he left 
with you ? are you not an adult yourself ?" But Tangiia replied, ** That 
will not do; I am going to set him up as an ariki." But tlie sister 
•objecting, argued with lier brother, saying, ** Don't do that! There 
are plenty of mataiapo [minor chiefs] , and priests suitable for the 
office." But the brother insisted that (the boy) should be an ariki. To 
this the sister (eventually) agreed. 

307. Then tlie sister Raka-nui asked, ** Who is he to be the ariki 
of ? " Tangiia replied, **He is to be ariki over Te Neke, Te Kaki-poto, 
Te Atu-taka-poto, Te Ataata-pua, Te Tavake-moe-rangi, Te Tavake- 
oraui-au, Te Tata-vere-moe-papa, Te Kapa-tavarivari, and Te Mana- 
une." [These were the clans, or divisions of Tangiia's people, and, as 
already remarked, they probably included many Melanesians taken as 
«ailors, slaves, etc., by the Polynesians. The Mana-une (or Mnna- 
hime) are known traditionally to the Maoris; the word means a 
•cicatrice, such as the Melanesians mark their bodies with. A tribe of 
that name lived in Mangaia Island, and also in Tahiti, while in 
Hawaii they were known as Mene-hune.] Tlie sister then approved, 
«aying, **lt is right!" She then asked, **What is his name?" to 
which the reply was, '* Taputapu-atea ! " To this the sister agreed ; 
and then new names were given to the lad, **Te Ariki-upoko-tini." 

[The chief of many heads, or ruler over many people] and Ta-i-te- 
ariki [a name borne not long ago by one of the West Coast, N.Z., 
•chiefs] . After this Raka-nui gave her canoe into the hands of 
Tangiia, saying, " Here is a second canoe for your vessel." When 
this canoe, named '*Kai-oi," was delivered to him, it was taken to 
the sea-sliore and served to convert the original vessel into a purua, or 
double canoe; the work being done by the navigator (and naval 
•expert) Tamarua-pai. When this work was completed, all Raka-nui's 
property was taken on board, and she, her husband Maa, and their 
•children joined Tangiia's party. In consequence of this joining of 



Digitized by 



Google 



MMory mud Tradtthn^ of RaroUmgu. 195 

"the two oanoes (to make it a double one—purua) Pai received the 
name of Purita. And now Taugiia and his sister broke out into song, 
and sighs [being somewhat anxious as to whether] death or salvation 
f laid before them]. 

308. [Tlie song will be seen in the original, but cannot be 
ti*anslated, though much of it is clear.] 

TU-TAPU IN CHASE OF TANGIIA. 

309. News was now received that Tu-tapu was near; so the 
•exi>edition left for Porapora [forty miles W.N.W, of Huahine, a 
somewhat low island with a beautiful precipitous peak on it] . On 
their arrival there they commenced to consecrate the ariki [Iro's son J, 
but tlie ceremony had not reached the stage of girding with the 
«carlet belt, when Tu-tapu arrived on the scene, in full pursuit of 
Tangiia. The latter then composed the following, expressive of his 
love [to Iro's sou ; for wliich see the on gin ill], 

310. The expedition now left Porapora and sailed for Rangi-atea 
£Ra4-atea Island a few miles S.E. of Porapora] , where the two 
vessels (of Tangiia and Tu-tnpu) closed, and passed along the coasts 
together, those on Tu-tapu's vessel shouting out, ** Give up my god! 
•Give up my god ! ** [It will be seen in paragraph 293 that 
Tangiia's messengers had taken Tu-tapu's god Bongo-ma-Uenga] . 
And now night fell, and a gust of wind [and ? mist] arising, which 
hid them, and their courses separated. 

311. Tangiia now returned to Fiji [a distance of 1,730 nautical 
miles, due east to the Lau Islands, to which no doubt, they went. It 
vrould not take them long as the trade wind would be abaft the beam 
aU the way] . On arrival all hands were ordered ashore the better to 
enumerate the crew (and passengers), and it was ascertained that 
they totalled two hundred men, whose place on board was to be the 
Jcaiea [the main one of a double canoe], while to the women and 
•children was assigned the ama [or lesser canoe, the word usually 
means an out-rigger]. 

312. At this time the following song was composed [which seems 
to be a sort of prayer for good fortune in the future] . And then this 
name of "the two hundred" wan given as a name for Tu-pa-moa- ariki, 
thus, "Te Rua-rau.*' [= two hundred, but all Polynesians counted 
men by pairs, so it means four hundred, rather a numerous crew for 
41 Pt)lynesian pdi^ or ship.] 

TANOIIA MEETS KARIKA. 

313. After the above tlie expedition visited many islands exhalting 
the fame of the vessel [ — its name was **Taki-pu"], even those 
islands towards the sun rise. When they were at the island called 



Digitized by 



Google 



196 JOURNAL if TNC FOLYNESIMM iOCMETY. 

liaketu ^ ^ they met (tlie Samoftu chief) Karika. On sighting thi» 
oanoe, Tangiia called out, ** There is a vessel." 8oTu-iti and Nukua- 
ki-roto climbed tip the mast to see it better. As soon as they discovered 
that it was Karika, they reported to Tangiia, saying, ** It is Te 
Tai-tonga ! [another name of Karika's'^J and you will be killed. **" 
Tangiia asked, 'Mlow many men has he?" *'A great many! A 
great many !" was the reply. ** What is to be done ? " said Tangiia. 
** What indeed ? There is only one course open, you must deliver up 
the rangi-ei (the liead plumes) on your head [i.e., must give up the 
supremacy to Karika]. 

314. Soon Karika's vessel drew near, quite close to the other^ 
and Karika came <m board that of Tangiia. Tangiia had already 
sent below all the able bodied men. leaving none but slaves, children 
and the decrepit on the deck of the vessel, so that Karika might not 
see them (the men). Tu-iti and Nukua-ki-roto now urged. '* Give up 
(your) ranyi ! Deliver over the supremacy ! " So Tangiia took off 
the red plume from his head and was about to hand it to Kaiika,. 
when Pou-te-are rushed up and knocked it out of his hand, and then 
took charge of it, clim])ing up the mast, and placed it on his own 
head. Then Pai [the navigator of the vessel] rushed after him and 
knocked it off Pui's head, and it fell down (? on the gunwale). 

11. Mr. Savage infonns me this is not Maketu — now called Mauke — of the 
Cook Group, but one of the Fau-inotu islands. It is not, however, shown in Mr. 
J. L. Young's list of those islands. See ** Journal Polynesian Society,** Vol. VIII.,. 
p. 264. In my opiiii(m the island is not in the Paumotu Group — properly so 
called — but is the island called on the charts Ma4tia, situated some 60 miles to the 
east of Tahiti. This island was formerly called Tuhua ; it is of volcanic origin, 
and is some 1,600 feet high. At the present time the island is named Me*etia 
(which is the same as the chart name Ma4tia), but was formerly also called 
Me^etu, but in consequence of the word tu in the last syllable entering into the 
name of one of the taptted chief's names, the last syllable was changed to xa. As 
the Tahitians do not pronounce the letter A% and as a and e are universally inter- 
changeable, we have the name Maketu, as the inland was known to the- 
Rarotongans. Thin is an interesting example of the letter changes in Polynesian 
languages ; it is more interesting perhaps to know that under its name Tuhua it 
was known to Maori tradition, and also as Maiteka in the Taranaki tradition of 
Wws *' Aotea '* canoe. See ** Journal Polynesian Society,'* Vol. IX., p. 213. The 
island is known to the Paumotu people as Mekiteka, and its European name is^ 
Osnaburg Island. I would remark here that probably this is the island known to 
Maori tradition in which a violent volcanic eruption took place some time before 
their ancestoi*8 left those parts in the fourteenth century. The name Tuhua, the 
meaning of which is now lost, evidently meant a volcano, or volcanic eruption 
formerly. Thus we have Tuhua island in N.Z. an extinct volcano, Tofua in the 
Tonga Group, Tofua-tuana*i in Upolu, Samoa, Tofua, Savaii in Savaii Island,. 
Samoa, all extinct volcanoes. 

12. Karika came from Manu'a Island in Samoa, where his descent is retained 
on the archives of that island. It is probable that a feud already existed between 
him and Tangiia. 



Digitized by 



Google 



History and Traditions of Rarotonga, 197 

Hence is tliis circumstance called " Au-topa " [the fallen phime, 
ov fallen supremacy]. 

315. Thoii Pai went on to the ova or sea-board of the canoe and 
«eized the emblem of peace ; this was the second tune the plumes had 
been taken ; hence is this circumstance galled '* Rua-kake-ovn." This 
is what happened when the rangi was [attempted to be] given to 
Karika at that time, at Maketu. 

316. And now commence<1 fighting; and Mu-tokerau and Mu- 
tonga L? uanies of winds, used emblematically for the fuiy of the 
£ght], came and fought for the supremacy, and evil nearly befell 
them tlirougli the paucity of men, for only the decrepit were [? on 
4eck] . The vessels were fastened close together ; the paddles were 
ji^eized, when Tangiia called out to his men, ^^ Hold fast you men, take 
the paddles, the whole of them." It was in consequence of his 
•expression taki-pu and taki-tumu that his vessel received the names 
^* Taki-pu " and "Taki-tumu" [which I think means "destroy root 
and branch "] . Fear now fell on all men [the enemy], and Karika 
-was in great trouble, whilst Tangiia's men felt their spirits rise. 
Each man now took to the paddles, and towed Karika's yessel from 
Maketu to Maiau. And so Karika decided to make peace with 
Tangiia, and [to cement it] gave him his daughter Mokoroa-ki-aitu 
in mariiage. And then the canoes separated. [The next sentence 
appears to indicate that this was at Ta*a, perhaps Taha'a near 
lla*iatea, but it is not clear.] 

TANGIIA SE'JTLES AT RAROTONGA. 

317. Tangiia had asked of Karika [who was a great navigator. 
Dr. W. Wyatt Gill says he made eight voyages from Earotonga to 
8amoa], saying, ** Where is Rarotonga?" Karika then explained, 
'* There." [Another tradition is more full than this (which appears to 
liave something omitted), and describes the course Tangiia was to take 
to tlie south and west to find Rarotonga.] The vessels now separated, 
Karika going his way, and Tangiia an his. But the latter missed his 
mark and reached those parts [in the south] whei-e are great waves 
and currents, and he thought, **This is the * Tai-rua-koko * " [place of 
monstrous waves, reported by Ui-te-rai>gi-ora, see par. 277, Part VI. 
liereof. Another account says they knew they were too far south by 
dipping their hands in the sea and finding it intensely cold]. . From 
{those parts) they returned north, and then discovered Rarotonga, 
:and finally reached the land [which was to be Tangiia's home ever 
afterwards. Another account says that the name of this beautiful 
island is derived from Karika's directions to Tangiia ; he was to go 
raro (west) and tonya (south). Here we may leave the expedition 
while the Sage describes the first inhabitants of Rarotonga, see 
Part VIII., and take up Tangiia's adventures later on]. 



Digitized by 



Google 



198 JWRMAL OF THE POLYME^AM SOCIETY. 

KA KOO A TANOnA-NUI I TAITL 
Kiia anan maira a Taiigiia, k<> Raiigi te iiigoa ta Ka'n-kura i topa 
ki toiia utaro. E kite akera a Pou-vauaiiga-roa i te tauia a te tuaiiie 
— a Kau-iigaki — kua rave aia nana ; kua topa aia i te iiigoa o tana 
tamaiti ra ko Tangiia-iiui. Kiia rave katoa maira i ta tetai tnaine 
tamaiti — i a Tii-tapu — ei tokoma i ana iiga tauianki nana ra ko Tii- 
tapu ; ko ta Kaii-ngaki la ko Taiigiia, ko ta te teina la, ko ta Maongo^ 
ko Tiitapu i'a. E riro roa rai maira na Pon-vananga-roa ana taninriki. 

279. Tei*a te aerenga niai o te kapionga ki a Tn-tapn : Ko Nga- 
npoko-tunia o Taie nia Tai-rann — 1« Ono-kura i anan i te vaine, i a 
Te Ata-nna. Anania e Nga-upoko-tnrua ko Nga-niani. Anau akera 
tana ko Kotukn-tea ; anan akera tana ko Pon-tea : kna rave aia i te 
tuaine o Pou-vananga-roa ei vaine nana, ka anan mai ei a Tn-tapn, i 
riro ei a Tu-tapn ei tnakana no Tangiia, 1 riro ei a Tangiia ei teina,. 
no te mea, e tuakana a Kan-ngaki, e teina a Maongo. Ko te ariki a 
Maono, ko te karakia te teina, a Ken. 

280. Teia te tnatna i tana an tamariki nana i*a, a Pon-vananga- 
roa ra. Kua akataonga aere aia i ana nga tamariki nana ra; ko 
Maono, ei ariki aia ; ko Tu-tapn ei ariki aia ; ko Tangiia ei tavana 
aia. Kna tupu iora te pekapeka i aia, koia kare i akaankiia. Ko 
Tu-tapn, e ariki la ki Iva. Ko Maono, e ariki la ki Taiti. Kua liri 
akera a Tangiia, kua ao atnra i te taonga i mna i te marae, i Avaiiia : 
kua tavaitai maira te ann nana, kika atura a Maono i loinga i te 
kainga, aere atnra ki te inaunga. Kua nro maira te kainga ma tetai 
pae i te kuru ki aia — ki a Tangiia. Kua peke a Maono i te nekeneke 
ki te tara-pakuivi ; ka tai te kino ka tupu. Kia tavaitai mai te 
taonga ma te kainga ki a Tangiia, kua tatari mai nga teina i tetai 
tapa kainga ma tetai taonga na i-atou, kia oronga mai a Tangiia no 
ratou; kua vai tei reira ei kino ki roto i a ratou. Teia tana aronga 
ra — ko Nuku, e Ai, e Ika, e Mere, e Uki, e Maraka. 

281. Teia te rua o te pekapeka i a Tangiia e Tu-tapu ; ko te vai, 
ko Vai-iria; e vai tapu no Tangiia. Kua tupu te tamaki ki tana vai 
ra, ka pa4 ai e Tu-tapu, ka tamaki ki reira, ki Patu-te-ere-tiki e 
Umu-toto-tatatou. Kua ati te nio o Tangiia, e kua ngaa te tia a Te 
Tua-ki-taaro. [Tera te aiteapga o Tua-ki-Taaro, ko Avaiki la, ko Te 
Tua-ki-taa-poto, ko tei kore ia i ngaa, e te aiteanga o Te Tua-ki-Taa- 
poto, koia a Iva e Taiti.] 

282. Teia te torn o te kino, ko nga ika e rua, ko te Karatea e te 
Onu. Kua manono mai a Tu-tapu, e tipu te mango, nana te mimiti, 
na Tangiia te iku; e vai te Onu nana tetai pae na Tangiia tetai pae. 
Kare ra e tika i a Tangiia, kua tupu iora te kino i a rana, ma te ke ki 
tetai ki tetai. Kua oki atui'a a Tu-tapu ki Iva, kua aere atura a 
Tangiia ki Mauke, ki nga tamaine a Te Tata-nru- ariki, ko Te Pua- 
ranga-uta tana vaine ; ko Pua-tara e Moe-tuna nga tamaine. 



Digitized by 



Google 



Hisioty and Trmdifions of RaroUHigm. 199 

283. £ tae atura a Tangiia ki Mauke kua kake atura ki uta ki te 
tapere ra, ko Utaki, ko Rangi>iuauuka te koutu. E kite atura aia 1 
aua nga vaiiie ra, te iioo ra ki ruuga i te renanga, te tukai ra i te ara. 
Kua akapiri aere atu aia iia roto i te iigangaere; e tae atura aia i 
vaitata, kua tae uiai te au ki aia, kua atu iora i te pee, na-ko atura : — 

Ka tukaiia, ka tukaiia tau oki, rirerire e — 

Ki te ara-kara ua, Pua-taru e, ki au e — 

Ka moe noa te vaine, e Moetuna, 

E tuugou o mata e te ariki 

Ka kite ra ia koe — e — 

Ka tukaiia ra, ka uka ia 

Ka ukaia iiiai ana. 

Kotai pakia ia, a paki akanui, 

Katakata noa koia'e — 

Kua ing^ te ta urua, Puatara e — ki au e, 

Ka moe noa te raine, e Moe -tuna, 

.A tiuig^u nga mata o te ariki 

Ka kite ra ia koe e, a kia mou a aroa, 

Akatae, akatae e rire taku au ki Mauke e — 

Ra Pua-tara ki au e— 

Ka moe noa te vaine Moe -tuna, 

Tungou nga mata ki te ariki 

Ka kite ra i a koe e — 

Ko akaa, akatere ki Mauke e — 

Ki a Tama-tino-rere, kua rere ua 

Ki te rangi tapu e rire e — 

Ko te rei o Tangiia, Pua-tara e, ki au e — 

Kh moe noa te vaine Moe -tuna, 

A tungou o mata ki te ariki, 

Ka kite ra i a koe e — 

E poupou pou ra ki tai e — 

Ko te tai o Kori-mata oki 

Ko te kau ra teua ra, 

Ki te tama rirerire e 

Ki te kape uti i a Pua-tara e, ki au e — 

Ka moe noa te yaine e Moe-tuna 

Tungou ng^ mata ki te ariki, 

Ka kite ra i a koe, e — 

E ko ikiiki iki na ruuga ki te tai 

Opukia ko te kauranga tena ra 

E te tama a ki runga i te koko 

Yl taku uutia ua, Pua-tara e — ki au © — 

Ka moe noa te vaine Moe-tuna 

Tungou nga mata o te ariki 

Ka kite ra i a koe, a kia mou aroa — e — 

284. E oti akera tei reira, kua oki attira ki Taiti. Kua ui maira 
te tuaine ki aia i te kuru. Kua karanga atura aia, " Kare au i aaki.'*^ 
Kua niaiiako aia e, tei i a Taugiia, kare aia i kite e, e na Tutapu i 
aaki, kua taria ki Iva. Kua tupu ia pekapeka i a raua ma te tuaine 



Digitized by 



Google 



200 JOURKAL OF THE POLYKESIAK SOCIETY. 

ki mua ki te marae ; kua tupu ia pekapeka ki reira, ki a Tangiia rai 
« Maa. Kua anna e Tangiia te tonga-ua onii ki te nma o Maa — ^ko 
to Maa puku la ki a Tangiia. Kua rave atura te tuaine i te vaka, i a 
*^ Kai-oi," kua akatei^e atura ki Uaine i kua aru atura a Maa i reira. 

285. Kua noo iora a Tangiia i Taiti, kua aro i te vaka, i a 
^' Tuna-moe-vai." 'Ko ta Ka'u-kura la ingoa i taua vaka ra; ko ta te 
tuaine ingoa, ko "Te-tika-a-te-tuaine" la. Kia oti te vaka kua topa 
a Tangiia i tana ingoa ko '* Taki-pu," kua akatere atnra ki Avaiki ma 
te pa-enua ravarai. E oki inaira ki Taiti kua unga atura i a Tino-rere 
ei tiki i nga tamanki ki Mauke. 

286. Kia oki mai i Mauke, tera a Tu-tapu kua tae mai ki Taiti, 
mai te tangata katoatoa. Tera to Tu-tapu aereuga, e tere tamaki, e 
a5 i te rakau, i a '^ Te Amio-enua," ki te inetua, ki a Pou-vananga-roa ; 
« te rara-roroa — koia te tangata. Kare roa akera i paria e Tangiia 
aua nga apinga ra e torn ; me te rakau, e nga ika e rua. Kua tupu 
iora te taumaro i a raua, e keta a Tu-tapu nana ; kare a Tangiia e pa. 
Kua akaruke i tei reira a Tu-tapu, kua pati ki' nga rara-kuru. Kua 
tuku atura a Tangiia i nga rara nona ei akamaru atu i aia : e kare 
rava akera i maru mai. 

287. Kua tupu iora te tamaki, kua koi iora a Tangiia i tona 
tangata, i Te Kaki-poto, e Te Atu-taka-poto, e Te Kopa, e Te 
Tavake-moe-rangi, e Te Tavake-oraurau, e Te Neke, e Te Ataata- 
apua, e Te Tata-vere-moe-papa, Kua mavete i teianei, kua taka a 
Tu-tapu ma tona nuku ki Tau-tira, kua taka a Tangiia ki Puna-auia 
ma tona nuku. 

288. Kua tangi a Tangiia ki te enua, kua tumu aia i te pee : — 

E tarina koia, i rimga a poi noa e — 

Ka rere po, rere ra e — 

Patiu te oro e, pua ara ua, 

Ko te mate e — 

Patiii te oro e, ka naku oi an e — 

Ko Taiti akarere e, ka rere e — 

E[a rere poi rei reira e — 

Patiu te oro e rai, 

Akua ara ua ko te mate e — 

Patiu te oro e rau akatea, 

Te mate noa ruru — e — 

E te pua ra oti ake ooia e— 

Ko te ua teia : — 

Ka kapiki ana e tamaki e — 
Ko tua rau enua ra 
Ko te aere mai nei, katoa ra, 
Te area tai e, ka rere poi rei ia. 
Pati te oro e, e pua rau ua, 
Ko te mate e. 



Digitized by 



Google 



Hiitoiy and TradiiionM of Rmrotonga. JMl 

Pati te oro e, ko te aere mai nei e, 

Ko te area tai oki» kua kapikipiki e — 

Ko iku i Taua-roa e — 

Kua topa a Vaimoko — 

Te aro aa ake nei te tamaki, 

A Taugiia, ki mng^ kau-aka ia, 

Ka rere po rei ia, 

Pati te oro e poa ran ua, 

Ko te mate e. 

Pati te oro e, ua a Tane nana e, 

Ko Akamatii kua e, kia mate 

Te marama rei iri e — 

Ko te kaa ia na te taunga, 

Na Keu-totoa e tataa ki Motu-tapu 

Ka rere poi rei ia, 

Pati oro e, e pua rau ua, 

Ko te mate e — 

Pati te oro e rau aka ia 

Te mate noa ruru e 

E te pua, oti ake ooi e. 

289. Kua ta te taiuaki, kua maata te kino, kua ki tetai pae i a 
Taitiitelva; kua peke a Taugiia. Te aere ua maira te Iva. Kua 
rave iora a Tangiia i tona au tangata ma te pal, kua tuku atura ki te 
moaiia, kua tari i tona apinga ki runga i te pai: i te atua ko Tonga- 
iti, e ko Bongo ma Tane, ko Kua-nnku, ko Tu, ko Tangaroa. Kua 
riro e rua i a Tu-tapu a Rongo-ma-Uenga, e Maini-mamao. Kua apai 
i te nooanga, i a Kai-au-unga, ki te pal. Kia ope te apinga ki te pal,, 
kua oki, kua ta i te taniaki ; topa atura nga tamanki ki te raaunga — 
ko Pou-te-anuanua e Motoro. 

290. Teia te mea i topa ai nga tamariki, no te kotinga i te tamaki,. 
i te ikuikuanga ki nga teina, ki a Uki e Maraka, ^'Ei Vai-tietie 
korua, e nga teina ! noo ei ma to korua tini." Ki a Ika e Mere* 
<< Ei Vai-pa-ekeeke korua e nga teina noo ei ma to korua tini." Ki 
a Nuku e Ai, " Ei Vai-pikira korua e nga teina noo ei ma to korua 
tini." Ko nga vai tamaki teia — I eke, i tauna ai te tuaivi, i mate ei 
tetai tamaiti a Tangiia i te ka ki te al — koia Pou-te-anuanua. Kua 
eke katoa ta Tangiia vai tamaki, kua topaia ki roto i te tai, kua ivo 
atura a Tangiia na runga i te arakaoa ki runga i te pal. E tae akera 
aia ki runga i te pal, kua akara ki uta ; kua ngaromia te enua i te 
aual, kua ka te tua-ivi. 

291. Kua akara katoa maira a Taa-kura, ki raro i te ai e ka ra. 
Kite iora i a Motoro i roto i te al ; kua kapiki maira ki a Tangaroa,. 
no-ko maira, " Aue ! teia ariki e, ka ka nei i te al e." Kua kapiki 
maira a Tangaroa, ki aia, '*E akapeea ! e atua koe, e tangata tera.'^ 
^^Eia! ka oro au ka tiki i taku tane." Kua akaunga atura a Tanga- 
roa i aia, na-ko atura, '^ E oro ki a Retu, kia oniai i te pakoko-ivi ei 
tapokipoki ei." Kua pati atu i tetai takao-matangi ei uuna i aia, kua 



Digitized by 



Google 



30a JOURttAL QF THE POLYHESIAK SOCIETY. 

orangaia maira te niatangi-nui ; kua na roto aia i relra ; kua tapoki 
iora, kua apai atiira ki A'lia'u ; na Te Muii e Te Pepe i apai atu. No 
reira i topaia ai ki a Tino-maiia. 

292. Kia oti te akaraaiiga a Taiigiia ki te enua, kua iriea iora ki 
te enua na-ko atura : — 

Ka aroa mea ra e, ki takn-enna, 

Ko Taiti ka vaio nei 

E[a aroa mea ra e, ko taku koutu, 

Ko Pure-ora, ka vaio iiei 

E[a aroa mea ra e, ko taku yai inu, 

Ko VHi-kara-a-mata, e Marama-ata-kai, ka vaio nei, 

Aku vai pai ko Vaiiria, ko Vai-te-pia, ka vaio nei, 

Aku tapere ko Puna-auia e Papeete, ka vaio nei, 

Taku maunga, a Tikura-marumaru, 

E Ao-rangi, ka vaio nei, 

E ae oki aku tamariki, 

Ko Pon-te-anuanua e Motoro i mate nei. 

Aue tou e ! aku tamariki e, 

Aku tamariki e, Aue tou e ! 

E Pou-te-anuanua e, Aue tou e! 

E Motoro e ! E Motoro e ! 

293. I muri ake i tei reira, kua akauiiga atura i a Tuiti e Te 
Nukua-ki-roto ei tiki i tetai iri, koia te iriiri i te marae, e poreo, e te 
tikoru. Taua apiiiga ra ei atua no ratou, Kare atura raua 1 aere ki 
tei reira ; kua aere atui*a ki te keia i a Hon go-ma -IJenga kia riro 
maira ki runga ki te pal. No aua nga atua ra i aru aere ua ai a Tu- 
tapu i a Tangiia. 

294. Kua akatere atura te pal ki raro i te pa enua rayarai, e tae 
na atura ki Avaiki, i te kai)*au aere i nga taniariki ma te enua. E 
mamae tu ke to ton a ngakau i nga tamariki. Ko Pai tei iri ki runga 
i taua aerenga ra, kua rave a Tangiia i aia ei toa nona. I aia te 
akatere i te pal ki Avaiki. Ko Teivao-ariki to Pai iiigoa i topaia ei e 
Tangiia. E tae atura te pal ki Avaiki kua eaau maira te pu e te pan, 
Kua ui maira a Tangiia, na-ko maira, " E pu ea? E pau ea ! e kaara 
ea?" Kua tuatua maira a Pai, '^ E pu akaariki ; e akaau takui*ua; 
e pau e akaau takurua ; e kaara, e akaau takurua ; e evatapu, e akaau 
takurua ; e eiva e akaau takurua ; e tei*e na te atua tini, e akaau 
takurua ; e ara-maora, e akaau takurua ; e ui-taura, e akaau takurua ; 
ko taua au mea ra i a taua e akarongo nei." 

295. Kua kake atura ki uta i te enua. E tae atura ki uta kua 
akaunga atui-a i a Te Mata-iri-o-puna kia aere i*aua ko Pai ki a 
Tu-te-rangi-marama, i a Rua-i-te-kari, i a Tu ; ia Tangaroa; ia 
Rua-te-atoDga ; i a Taakura ; i a Tupua-nui ; i a Kangatira ; i a 
Tu-vaero-kum ; i a Te Kura-akaipo ; i a Aronga ; i a Tu-avaiki ; i a 
Tu-tavake; i a Kongo ; i a Uenga ; i a Maru-mamao ; i a Kau-kura; 
i a Tonga-iti ; i a Nge-mua ; i a Mumaiio. Tei te akaau takurua 



Digitized by 



Google 



History and Traditions of Rarotonga, 203 

Himke to ratou uipaimga i teiaiiei ; ko te tuatau la i tae atu ai a Taiigiia 
ki Avaiki inei TaitI atn nei, ka oti ei te tamaklanga a Taiigiia ma 
Tutnpu i te luateanga o Pou-te-aiinanua ; ko te mua la o te tauiaki- 
anga. 

296. Kua ai iiiaira a Tu-te-rnngi-iuaraiiia i toiia aerenga, na-ko 
4itni'a, '* E aa koe i topa mai ei ki Avaiki iiei ? " Kua tuatua maira a 
Taiigiia ki aia, " E oru-kanga toku aerenga ; e ta inaua ma Tu-tapu. 
E kua luate aku tamariki, ko Pou-te-auiiaiiua e Motoro; e kua peke 
a,a i Taiti e te au mea katoa, kua ope. E teianei, kare ua akn ravenga." 
Kua ni maira taua ^ronga katoa ra, iia-ko maira, '^ E aa te ara i ta 
korua taniaki?" Kua aaki atu)*a aia, na-ko atura ra. '* Ko uga 
taoiiga; ko te ariki ; ko iiga ika ; ko te onu, ko te rarntea ; ko te 
rararoa ; ko te i*akau — ko Te Amio-enua ; ko iiga (ma la i ta niaua 
taniaki." 

Kua karanga maira a Tu-te-rangi-iuarama, na-ko maira, *' Mari 
rai koe kua tapu, kua kore e pa." Kua karangaranga iora ratou, 
ratou ua-o-rai, na-ko akera, ^' Ka akaniana tatoii i aia, e ka akataka i 
tetai enua nona." Kua kapiki maira a Tonga-iti ki a Taiigiia, na-ko 
maira, **Tera to taua enua — a Tiiinu-te-varovaro — ka aere koe ki 
reira, e ki reira koe mate ei." E kua tuku maira i te mana nona — 
koia te ira o iiga atiia. Tera te aiteanga, ka aii-tu-roa aia i te ta- 
makianga katoa, ka riro ra te re ki aia. 

297. E muri ake i tei reira kua pati atura aia i te au atua ravarai, 
kia uta ratou ki runga i ton a tere. Teia taua aronga ra, i to ratou au 
uuku nia to ratou au tere : Ko Tangaroa, ko tana tere ma ton a au 
nuku. Ko Tu-te-rangi-niarama ko tana tere ma tona au nuku. Ko 
Tu-tavake, ko tana tere ma tona au nuku. Ko Bua-i-te-kari, ko tana 
tere ma tona au nuku : Ko Kau-kura, ko tona tere ma tona au nuku : 
Ko Mumai-io, ko tana tere ma tona au nuku — (koia a Tou-tika i 
muringao e tumu-toa, e tapakia, e uka, e veko, e maai-a-nuku, e maai- 
4t-raiigi ; ko te aiiaunga la a Mumai-io, tokoitu ratou — tokotoru 
tamaroa, tokoa tamaine). Ko Rua-te-atonga, ko tana tere ma tona 
eu nuku; Ko Ai*i, ko tana tere ma tona au nuku ; Ko Tupua-nui, ko 
tana tere ma tona au nuku. Ko Taa-kura, ko tana tere ma tona au 
nuku. Ko Te Kura-akaipo, ko tana tere ma tona au nuku. Ko 
Tongaiti, ko tona tere ma tona au nuku. Ko Te Hangi-puta-rua, e 
Ai-inarie ko ta raua tere ma to raua au nuku. Ko Maru-mamao, ko 
tana tere ma tona au nuku. Katai aua mea ka rauka i aia i reira. 

298. Kua pati akaou rai aia i te takurua ma te au mea ravarai o 
te takurua. E kua tika, kua riro maira taua au mea ravarai nana. 
Teia taua au apinga ra : Ko te eiva: e nma tu o te eiva; ko te peu — 
koia te eiva— e kaara, te akaingo ; ko te puapua-aki — koia te tioi — e 
pau tona akairo ; e tutauniaa, ko te pokara tona akairo ; ko te kapo- 
rakau, e kaara tona akairo ; na Tu-tavaki la eiva ; ko te eva-tipa — 
koia te kopuporeo ki mua i te kopu, e te kopu-porio ki nga rinia; e te 



Digitized by 



Google 



204 JOURNAL OF THE POLYMESIAH SOCIETY. 

papata ; e te tau-mua-poupou ma te peketue ki runga i te taukupu, » 
te pare-eva-tipa ki rnnga i te upoko, e te rauini ki muri i te tua, e te 
parae tia ki inua i te rae, e nga takakaravia ki nga tannga e te takumi 
ki te kautuai, ko te tua-a-maim ki muri ei akamaiiea la a muii i te 
pare. 

299. E taka maira tauii au mea ravarai ki aia, kna akaimga i a 
Taote, Mata-iri-o-puna ei tari i te pu e te pau e te kaara katoa, ma te 
au mea katoatoa. E kua tono i a Tuvake-oraurau ei apai i te aiiga 
mei te takurua — koia te ara o nga atua, e te kaa ki A vaiki. Ko Te 
Mata-iri-o-puna, ko te pau ia a Tangiia, ko te anga e te akaau takurua^ 
ko Anga-takui-ua la. Ko Te Avaro no Rangi-raro aia, nana nga eau 
pu i apai ki runga i te paT. Kua uri i te tumu, te tere a Tangiia mei 
raro, mei Avaiki, mei te urunga, mei te papa, mei te uuga mei t& 
potiki, i topaia e, " Ko te tere a te tnmn." 

300. Kua oki maira te pal, kua tapae ki Ilea ; kua rutu te pau ki 
reira, kua akatangi te pu. Kua topa ki reira tetai pu i a Te Avaro ; 
kua aere maira te pa! mei reira niai e tae mai ki Kuporu. Kua rutu 
te pau ki reira, kua akatangi te pu. Ina ! kare ua te pu, kua topa ki 
Ilea i a Te Avaro. Kua oki atura te pal ; te tiki akaou ki Uea. E 
liro maira, kua piii mai a Katu i reira. E tae maira te pal ki Kuporu^ 
kua rutu te pau, kua akatangi te pu ki reira ; kua tumu te peeki reira,. 
kua na-ko akera : — 

Tela te pau i runga i a Atea 

E karu pau no te kotinga ia 

Te Tamoko-o-te-rangi, ka paupau ra nga vaka e, 

E mate i to akania e 

E Roiigo e ! kua oro ki Vai-nuku 

E ka paupau ra nga vaka e — 

I tuputupu ra i Avaiki e — i Avaiki mai ana, 

E papa ra, ko Akaotu, te etu taki riko e — 

E te purotu tama na Te Puta-i-ariki, e ariki e — 

Rutil i te pau i Kuporu e — 

Akatangi te pu mei Uea e — 

Ko te ape oki na te ariki Maro-kura ie ra, 

Ka paupau ra nga vaka e — e mate i te akama, 

E Kongo e ! kua oro ki Vai-nuku e — 

Ka paupau ra nga vaka e — 

I tuputupu ra i Avaiki e — i Avaiki mai ana, 

E rakau ko Te Tamoko-o-te-rangi e — 

I kotia ia e te ariki, e Ui-te-raugioi-a, 

Ei kiato, ei ova, i to pal, ia Tei Ivi-o-Atea nei, 

Ka paupau ua ra nga vaka e — mate i te akama oki, 

E Kongo e ! kua oro ki Vai-nuku e. 

Ka paupau ua ra nga vaka, ka paupau ra, 

Nga vaka oki, e rue e — 

301. Kua aere atura te pai ki Iti ; tei reira a Iro. Kua ui atura 
a Tangiia ki a Iro, " Tei ea te tamaiti ei ariki noku? Kua mate aku 



Digitized by 



Google 



Hisiory and Trad/Hens of Ranoionga. 205 

tamariki." Kua akakite niai a Iro, '* E tei Bapa roa ai ; kua aka- 
nooia e au ki jreira." Kua karanga atura a Tangiia ki a Iro, '^ Ka 
aere au ; ka taoi, ei ariki uoku." Kua akatika maira a Iro, na-ko 
inaira, ** E aere ! taoiia! " 

302. Kua aere atura te pal ki Rapa-nui ; kua kite atura i taua 
tamaiti ra — ia Taputapu-atea, tei ruuga i te akau ; te maruku ra i te 
pao i reira. Kua ui atura a Tangiia, na-ko atura ki aia, ** Koai 
koe ? " Kua tuatua maii*a aia, na-ko maira, '^ Ko Taputapu-atea 
au ! " Kua kapiki a Tangiia, na-ko atu ra, *' Aere mai, ka aere taua ; 
•e tiki au i a koe ; ka taoi au i a koe ei ariki noku.*' E riro maira 
taua tamaiti ra ki runga i te pal, kua aere atura te pal. 

303. E tae atura te pal ki Morea, kare rai a Iro i tae mai ; kua 
akakite atura ki reira e, " E kia tae mai a Iro e akakite atu kotou e, 
kua taoi a Tangiia i te tamaiti ei ariki nona." 

304. Kua oro maira te pai ki Uaine. E tae atura ki Uaine, kua 
•eke maira a Maa, e kite maira i a Tangiia ; kua kapiki maira aia, ki 
a Tangiia — kua tuatua maira, ** E oki ! kua tamatemate te vaie a to 
tuaine — a Raka-nui — e kua maro ; ei tan i a koe.'* Kua na-ko atura 
a Tangiia ki a Maa, ** Kare au e oki, e taunga ko Unutea, e purii 
ko Maro-tai, ka aere rai au ki uta." E tae atura aia ki raro i te ava, 
kua eke maira a Baka-nui ma nga tamaine, kua amuamu maiva i aia, 
na-ko maira, **E te tuna e! upoko nui, takiritia, topa ki raro, e ketu 
ana i te vai, e ketu ana i te vai. Ina ! e taku tungane ! kua takina 
mai na koe ki toku ngaii (? ngai), ki te moana nei. Ko te amoinu 
tena, E taku tungane ! o te kino." Kua tuatua atura to runga i te 
pal ki te amuamu mai ki to uta, ki te Papaka-tueraera, na-ko atura, 
^* Naau i taki mai nei ki Uaine e Ngangaua. Ko Uki ma Raka-mea, 
ka noo i te vaine, i a Raka-nui, i anau ai a Tarionge, ma Te Ariki- 
oa-ngata, e ketu mai i tetai para i Taiti. Ko au rai ia, ko Tangiia, e 
taunga ko Unu-tea, e purii ko Maro-tai. E vaka mania taku i tuku 
atu ki a koe, a kua paraaraa te ara ki a Uaine, ki taku tuaine, ki a 
Raka-nui. Oatu au ki uta. Tena a Keu tei uta.'* Kua aue iora te 
tuaine ki runga ki a Tangiia, kua oongi iora, aere atura ki te kainga. 

305. E tae atura raua ki roto i te are, kua tau iora i te umu-tara- 
kai na Tangiia, ma te aronga-a-vaka. E pangia akera ratou, kua ui 
maira Raka-nui, ki te tuugane, ki a Tangiia e, '* E, e aa koe i roaai? 
E aa te tupu i a koe ? E tei ea o tatou kainga ? Tei ea o taua rara- 
kuru? Tei ea o tatou atua? Tei ea te rakau — a Te Amio-enua? 
Tei ea nga tamariki? Tei ea to taua metua? '* Kua karanga atura 
a Tangiia ki te tuaine, na-ko atura, '* I roa au ki te tere ua i te 
moana, i te kaoe aere i taku riri ; ki te akaevaeva i aku tamariki." 
Kua na-ko maira te tuaine, '' Kua mate o tamariki?" **Kua mate 
a Pou-te-anuanua e Motoro ; kua riro te enua ; kua riro nga rara- 
kuru o taua ; kua riro tetai atua." " Tei ea te riikau ? " " Teiia." 
-** E aa ra te mea i roa kitu ua i a koe ? " Kua akakite atura aia e> 



Digitized by 



Google 



206 JOURKAL QF THE FOLYKESIAM SOCiETY, 

'* I roa rava au ki Ayaiki-te-varinga." *' A, tae roa koe ki Avaiki ! B 
aa te tuatua i Avaiki ? " '^ Tera te tuatiia i Avaiki, taku i kite ki 
reira; e taknrua i toku taenga atii; te akaau takurua ra, ma te rave 
i te au i>eu tu matini." 

306. Kua ui maira te tuaine ki a Tangiia, iia-ko maira, " E,. 
tena taua au mea ra kua taoi mai e koe ? " Kua akakite atura aiaki 
te tuaine, **Teia ravarai ; kua taki maira i te tumu, kare e toe.'*^ 
Kua karanga maira a*Itaka-nui, ** Tei ea te eva-tipa ? " "Teia ! '^ 
^'0! ka kite tatou ! " Kua tapu maira te tuaine, *- Ka uoo taua i 
kuuei ; ei tai tara koe i te euua, ei tai tara au." Kua karanga atura 
aia, ^' E ia, kare au e noo ; ka aere au. Tera toku enua ko Tuniu-te* 
varovaro ; kua akakiteia mai e Tonga-iti noku. Ka aere au ki reira 
e mate ua atu." ** !l^oai ia enua ? " ** E koai oki ! kare ua au i kite.'^ 
Kua akakite atu ki te tuaine, '* E tena ake te tama a Iro kua taoiia 
mai e au ei ariki noku." **Ei aa i vao rai ki a koe; kua tangata 
metua koe." Kua na-ko maira a Tangiia, **Eia, ka aere rai au ka 
iki ei e ariki noku." Kua tamaki maira te tuaine ki te tungane, " Eia, 
E Tama ! e manganui te taonga, e luataiapo, e taunga, e kaa." Kua 
maro rai te tungane ei ariki rai. Kua akatika maira te tuaine. 

307. Kua ui maira te tuaine ki te tungane, *' Ei ariki aia no te 
aa? " Kua akakite atura aia, na-ko atura aia, ^^Ei ariki aia no Te 
Neke, no Te Kaki-poto, no Te Atu-taka-poto, no Te Ataata-pua, no 
Te Tavake-moe-rangi, no Te Tavake-oraurau, no Te Tata-vere-moe- 
papa, no Te Kapa-tavarivari, no Te Mana-une." Kua akatika maira 
te tuaine, na-ko maira, ^^ Kua dka ! " Kua ui akaou maira te tuaine, 
ki a Tangiia, na-ko maira, '' Koai tona ingoa? " Kua akakite atura 
a Tangiia, ** Ko Taputapu-atea." Kua ae mai aia, " Koia ! " Kua 
tapa iora i te ingoa, ko Te Ariki-upoko-tini, e Tai-te-ariki. E oti ake 
tei reira kua tuku maira i te vaka ki te rima o Tangiia, na-ko maira, 
** Tera te rua i o vaka ki to rima." Kia riro maira te vaka — a 
** Kaioi " — ki te rima o Tangiia, kua apai atura ki taatai, kua purua 
iora i nga vaka ; kua aau iora a Pai i te vaka. E oti akera kua tari 
maira to Kakanui apinga ki runga i te pai ma tana tane e te tamaiti. 
No reira a Pu-rua i topaia ei i a Pu-rua ki a Pai. Kua eeva iora 
raua, raua ua-o-rai, i te irieaanga ki te ora, ki te mate. 

308. Kua na-ko maira : — 

Ka^u-kura taku tupuna e — 

Eia mai to tika, te tama e — 

Te pa tuki-papa ki te ipo e — 

Ko Ka'u-kura a ariki, 

Ko tei anake nei e — 

Te pa tuki-papa, tiria io i runga e — 

E Kongo, E Tane, Rua-nuku, 

Ko Tu, ko Tanguroa, rei iri e — 

Ko te Ta-ei-raugi, kia rongo Tangiia, 

Eia tuku ua ra, tuku ua ake e — 



Digitized by 



Google 



Hhiory and TradHhn$ 6f fhrotonga. 207 

T« pa tiild-pftpa ki te ipo e — 
Eo KH*a-kar» ariki, 
Ko te anuke nei e — 
Te pa tukipapa run e. 

309. KuR ee inaira e, teiii n Tu-tapii; kua oro atura ki Porapora. 
E tae atura ki Porapera, kua iki iora i te ariki ki reifa. Kare i 
tatuaia te maro-kura kua rokoia mai e Tu-tapu. Ka aruuiakina ra e 
Tu-tapu. Kua tupu iora t© aroa ki reira, kua na-ko maira :— 

£ ariki iki ua ki Porapora e — 

Ka tuaru-makina e, pati te oro e — 

Kna reka ia i ta oi rare e^- 

Ko te ariki ka toaru-makiim e, 

Pati te oro e, okotai koe, okotai au, 

Oro UH atu iia koe ki Ayaiki 

E ka tnaru-makina e, 

Pati te oro e, tapaiia apai e, 

K& tiiHru-makitia e 

Pati te oro ran aka ia, te mate noa 

Rum e te pua, oti ake io e. 

310. Kua akaruke inaira i Porapora, kua oro atura ki Rangi- 
atea, kua kapiti nga pal i reira, me to Tangiia e to Tu-tapu, i to 
tauaiii aereauga i tei reira ugai moana, ma te kapiki aere atu to 
Tu-tapu pai ki niua i to Tangiia e, ^'Eomai taku atua ! E omai 
taku atua ! " E tei te poirianga iora, tuku maira te takno mataiigi 
ei uuna i a ratou — taka ke atura. 

311. Kua oki akaou atura ki Iti-nui. E tae atura ki uta kua 
titiri atura i te taiigata ki uta, kia taka meitaki te raingaote tangata, 
kua akakatoatoaia iora e rua rau tangata, kua tuku oora (? iora) la 
ki a katea; ko te vaine nia te tamariki kua tukua la ki a ama. 

312. Kua tumu i te pee ki reira : — 

Ka uia katoa e, e rire, 

Ko te vaka ia uo Tangiia e— e riri e — 

£ mii, e anau e, te metua vaine, 

Ko na te tokerau ooki ana 

Te omai ake tai aroa, e riri e — 

Ka uiiiia ra ki Iti-nui, 

£ kia taka meitaki te rua rau tangata o Tangiia 

£ anau, e niii, e anau e, te metua vaine e — 

Kona te tokerau ooki ana 

Te oumi etai aroa, e riri e — 

E mii, e anau oki, e rue e— 

Kua tapa iora i taua rua rau tangata ra, i topaia ei ingoa no Tu-pa- 
moa-anki koia a *'Te Kua-rau.'' 

313. E muri ake i te reira, aere atura te pal ki runga, ki te pa 
enua ravarai, akateitei itiva atura i te pal ki te itinga o te ra. E tae 
atura ki Maketu, kua aravei atura ma Karika ; e kite atura a Tangiia 



Digitized by 



Google 



208 JOURNAL OF THE POLYUgSIMM SOCIETY. 

i a Karika, kua karauga atura ki te tangata, ^'E pa! tela!" Kiia 
kake atiira a Tu-iti e Te Niikua-ki-roto kl riinga i te tira, kua akara. 
E kite atura k<> Karika, kua akakite inaira ki a Tniigiia e, ^' Ko te 
Tai-toTiga teia ! kua mate koe ! " Kua ui atura a Taiigiia ki a raua, 
'* E ia ona tangata ? '' ** E rai ! e rai ! " ** E, ka akapeea ? '* " E 
ka akapeea oki, ko te rangi ei i to upoko taau e tuku.'* 

314. Kua vaitata mai te pai o Karika; e piri inaira ki te pae, 
kua kake a Karika ki ruuga i to Tangiia paT, Kua tapoki ioi*a a 
Tangiia i te tangata ki raro i te pal, kua vao uaorai i te unga ma te 
tamariki, e te tangata a[>ikepike ua ki runga ua i te oropapa i te pai, 
kia kore a Karika e kite. Kua raurau nmira a Tu-iti e Te Nukua- 
ki-roto, **E, tukua te rangi! tukua te rangi!" Kua kiriti akera a 
Tangiia i te Au-kura i runga i tona upoko, ka tukua ki a Karika. E, 
e kite akera a Pou-te-ari, kua oro maira aia, kua patu i taua Au-kura 
ra i te rima o Tangiia, kua topa iora ki raro, e riro maira i a P<m-te- 
ari, kua oro atura aia ki runga i te tira, kua pare ki runga i tona 
upoko. Kua oro atura a Pai, kua patu i taua pare rai i runga i te 
upoko o Pou-te-ari, kua topa iora ki raro — koia ** Au-topa." 

3 1 5. Ko te oronga a Pai i runga i te ova i te pai ka arapaiia ; ko 
te ruaanga ia raua i te mou i te an, ko " Bua-kake-ova " la. Ko te 
tu ua la o te rangi i tukuia ki a Karika i tei reira tuatau ki Maketu. 

316. Kua tupu iora te tamaki i reira; kua rere mai a Muu- 
tokerau, e Muu-tonga kua rere iora i te okoitu, ka tamaki no te au ; 
i tatakinoia ra e te kore tangata i te pal, e apike anake. Kua aau 
iora i nga vaka, kua kapiti kia piri ; kua mou i te oe, kua kapiki iora 
a Tangiia ki te tangata, na-ko maira, '^ Ka takipu ua mai kotou, ka 
mou to oe, takitumu ua te oe i te vaka." No reira i topaia ai taua 
vaka ra i a ** Takipu " e '* Takitumu." Kua io ra te mataku o te 
tangata, kua tumatetenga roa akera a Karika, kua rere te mauri i te 
maruaanga i te tangata o Tangiia. Kua mou iora te tangata i te oe, 
tanao, tauao, kua oe atura, e riro maira to Karika vaka i te taoi mai. 
E takipu mei Maketu mai e tae mai ki Maiau. Kua akaau maira a 
Karika ki a Tangiia, kua omai i te tamaine, i a Mokoroa-ki-aitu ei 
vaine na Tangiia. Ko te mataraanga i nga vaka, ko I'aa la 
{? Taanga). 

317. Kna ui atura a Tangiia ki a Kariki, na-ko atura, ** Tei ea a 
Karotonga? " Kua akakite maira a Karika ki aia, ** Tera ! " Kua 
matara nga vaka, kua aere a Karika i tona aerenga ; kua aere maira 
a Tangiia i tona aerenga. Kua taveva ke atura ki te ngai i uri-nui 
ei te au ; te na-ko aia e, ** Ko te tai-rua-koko I " Kua oki maira mei 
reira mai, kua kite maira i a Harotonga, e tae maira, e vaitata ki te 
enua nei, kua uru maira ki uta. 

(7m£ in te roaanga tei te SS2J) 



Digitized by 



Google 



THE PAUMOTU CONCEPTION OF THE 
HEAVENS AND OF CREATION. 



With Notes by J. L. Young. 



[Through the courtesy of Mr. Young we are enabled to reproduce 
a quite unique drawing made by Paiore, a man from the Paumotu 
Group, in 1869, representing the world, and tiie heavens above as 
conceived of by the branch of tlie Polynesians to which Paiore 
belonged. 

The photograph from which tlie picture is taken is in two parts, 
each nine by seven inches, and from this a tracing was made for 
reproduction on a smaller scale. The photograph is too faint to admit 
of a further reproduction by that process. The drawings of the 
canoes in the second photo will be reproduced hiter on. 

It is much to be regretted we have no further particu1a]*s as to the 
meaning of the vaiious parts of the picture, beyond what Mr. Young 
has supplied, fi>r no doubt, in former times the natives would have 
been able to explain the whole, but now alae! there is no hope of this. 

It would appear that in the Paumotu belief there are nine spheres 
in the heavens above, whereas tlie Maoris have twelve, and other 
branches eight. It is peculiar too, that the Paumotu people recognise ' 
separate names for the two sides of the heavens, or, what is more 
probable, that the name.s written on the right hand side, refer to 
spheres or periods of development of the earth, and that these names 
should properly be written on the nine stages so encumbered by figures 
that the native drauglitsman could not find room for them in their 
proper places. None of the Paumotu names for these spheres can be 
identified with those of the Maoris, though both peoples had the same 
ideas as to concentric spheres above the earth. Mr. Young's notes 

follow.] 

I was unable on my late visit to Tahiti to sight the original 
drawing, which has disappeared because of the death of the foimer 
owner. 

But ivom notes which I made at the time when the photograph 
was taken in 1892, and after consultation lately with certain aged 
natives of Tuamotu, I have arrived at the following as the probable 



Digitized by 



Google 



210 JOURNAL OF THE POLYMESIMH SOCIETY. 

text of the iiiscnption wliicli in the pliotograph before me, as in the 
one in your possession, is illegible in parts : — 

" Te man parau iiohoa i faaitehia e te feia tahito. E hohoa no 

te ao atoa nei e te parau atoa na te liua raa, parau oia nana e 

faiii te man uri. Tirara parau man pona tianiu no te api ru 

ati tune, ati tohe, ati niau ni, te fanaua uii a u na att buaka, 

a hio i roto i te pipiria nei ia ite e papai.'* 

Na Paiore, 1869. 

Which with the assistance of the old men aforesaid, was trans- 
lated as follows : — 

** The likeness (or description) of tilings made known to the 
people of ancient times. The form of this our World and 
the account of our ancestoi*s, and of the beginning of the 
movement of animal life. This is the true and succinct 
description (literally a bundle tied up with a knotted string) 
of mankind which was confined in narrow spaces, and of the 
origin of things and of the various trees (or vegetation) and 
of the bringing forth of animals which suckle their young,, 
such as four-footed animals. These are to be seen in this 
sheet of paper as understood by the writer, I, Paiore, 1869." 

The foregoing is written in the Tahitian dialect, but with an 
admixture of words in Tuamotu. The writer seems to have put *'i " 
«.nd " a " instead of " e " in several instances. 

I am not quite satisfied with the correctness of the translation as 
for example, '' efaui te mau uri'* — ''the beginning of the movement 
•of animal life" : and ** no te api ru ati tane," " of mankind which was 
•condned in narrow spaces " : but such was the interpretation of the 
old men referred to. They also held that the words "mau ni" 
literally '' all the coconut trees" means in the context, '' all trees and 
Tegetation," 

The woi'd "buaka'* — npig; was used by natives to signify any 
iour-footed animal except a rat. It is said that the drawing was made 
with a stick In the sand by Paiore, an aged tnhunga — wise man, who 
lived at Takaroa Island, but who was a native of Anaa, and that the 
^drawing was copied by a young relative who had been to school at 
Papeete: the inscription it is said was dictated by Paiore in the 
Tuamotu dialect, but written by the young man in Tahitian. 

NOTES ON THIl DRAWING. 

It is explained by the natives that tl»e lowest division represents a 
period when the world was inhabited by animals not known to the 
Islanders, and when the sky hung low over earth and sea. In the 
ihird division ai'e shown the first homicide, the fii-st burials, and tbid 
first canoe. 



Digitized by 



Google 



Digitized by 



Google 




:v-V:>:i^.-!°^^--^^- 



"^.'..'o -r.. 






%':\''-yi.: 



f 



...osa^o*^^^ 



OD 

z 
n 

o 

o 



O 

Z 

o 



.)J<PT**5|^i^ 



*J 



-^""^^^ 



1^^ 



^r^$ 






Digitized by 



Google 



The Faumotu Conception of the Heawone and of Creation. 211 

111 the fourth tlie fii*8t coconut tree, pandaiiiis tree, Puatea {Tourney 
for tin) tree, and Ton (^Cordia Suhcordata) tree, &c., &c. 

In the ninth, ou riglit side, the Oonstellatioii Scorpio, and on the 
left Bide, the moon and a star and a man making an offeinng or 
sacrifice at a fire. The names written outside each circle are as 
follows : — 



Outer Circle left side 


Tui ni ao 


Second 




Tia ruga a taha 


Third 




Gkru te fatumoana 


Fourth 




ragi te ke 


Fifth 




Tikohu ariki 


Sixth 




Turi hono 


Seventh 




Te tumu o Kuporu 


Eightli 




Tnre Ora 


Ninth 




Peaha 


Outer Oiixsle 


right side 


Kagi no 


Second 




Tumu no 


Thiid 




Tuniu haruru 


Fourth 




Tapatapa i aha 


Fifth 




Te piu honua 


SixUi 




Matau hiti 


Seventli 




Onovaru 


Eightli 




Kororupe 


Ninth 




Tuarati 



Tlie canoes* are those built by Rata ; the inscription above tlie sail 
of tlie middle one is ''Tutu nei tere a tetoira" — '^ Sailing with a 
light easterly breeze to tl^ westward." Note the ladder- like masts 
which were those used on the Tuainotu sea-going pahi canoe of old. 

I am doubtful as to the value of this drawing, as I think that the 
author was probably influenced by missionary teachings and by pictures 
which he had seen. 

For instance, the figures of the Assf and the Apef in the lowest 
division : the homicide in the third suggests Cain and Abel : the 
offering (if it was intended to signify such) in the ninth suggests 
Abraham and Isaac. 

Nevertheless, the general idea is that of ancient tradition : the 
raising of the heavens by human or rather superhuman effort. The 
names of the heavens are also interesting. 

♦ The drawings of the canoes will be reproduced later on.— Editob. 
t It is suggested that these figures represent a pig and a dog. — Editob. 



Digitized by 



Google 



TRADITIONS AND LEGENDS. 



COLLECTED FROM THE NATIVES OF MURIHIKU. 

(Southland, New Zealand.) 



By H. Bea'itie. 



PART XL 

IN common with their northern bretiu'en the Maoris in the south 
were exceedingly prone to superstition, and the collector of 
these traditions came across many instances of tliis characteristic. 
Several parts of Otago seemed to be specially liaunted by ogres, 
judging by the nomenclature. Some hills behind Waikouaiti are 
called Pukemaeroero, while the hills behind Tautuku and the 
mountains to the west side of the lower end of Lake Wakatipu, also 
bear the same name. The Maeroero was a wild man of the woods, 
covered with long hair, and possessed of great strength and craftiness. 
The name Maeroero also seems to have been sometimes applied to 
elves and fairies, but the word usually implies monstrous beings, 
whose existence was implicitly believed in up to quite recently. One 
old Maori said to the collector: — "The Maeroero which lived in the 
Owaka forest was a fearsome creature. Two knobs iu the Ratanui 
Range are called Puku and Miki after two sisters who were married 
to Te Waka-tau-j)uka, an uncle of Tu-hawaiki (and incidentally an 
uncle of the narrator also). One of these women wandered into the 
bush where she killed a kakaruai (robin), and the Maeroero pounced 
out and carried her and the dead bird away. She came back about a 
week later but was unable to relate her experiences owing to fright 
and collapse. Her friends made an unm (oven) and covered it with 
clay upon which the woman was laid. This act of tao-tchakmnoe was 
done to remove the spell and avert evil. * It shifted the tapti off her 
and made her an ordinary w^oman again,' '* concluded the narrator. 
The mention of tapu recalls the fact that the late Rakitapu of 
Port Molyneux once had a pig which he named after Hau-matakitaki, 
a mountain near his birth-place at Lake Wanaka. Then some one 
recalled the fact that that mountain, and mount Kaki-roa near it, had 
been named after ancestresses of his, so the pig became tapu^ and 
the}'^ could not eat it. A year later when the pig was in prime 
condition the elders consulted together, and decided that the naming 



Digitized by 



Google 



Trsditiw$ and Legends. 213 

of the pig }iad been done unwittingly, and that the m&na acquired 
from the ancestral spirited had all gone to its liead bo the head was 
carefully severed, and the rest of thiB animal was eaten. 

SAWDY AND THE MAEROEBO. 

In some notes which Mr. James Cowan kindly gave to the collector 
to make use of as desirdd, occurs the story of Sandy's shooting 
expedition. The collector Was also told this tale, but as Mr. Cowan's 
account is fuller it is herewith presented : — *' In the very early days 
Sandy, a pakeha, met a Macro on Puke-maeroero ('Fairy Hill' — ■- 
between Tautuku and Hakopa). The Maoris warned him that there 
were fairies there, but he would go. Into the dense forests he went 
shooting and firing at kereru (or kukUy pigeon). He hit a kereru and it 
dropped flopping to tlie ground. Just as he went forward to pick it 
up, suddenly a terrible figure — a wild, hairy man of the woods- 
appeared and menaced him. Sandy immediately rammed down a 
charge into his gun and hurriedly fired at the Macro at very close 
range. The shot should have struck him, but he was no mortal man 
to be killed by shot. He laughed loudly, snatched up the pigeon 
and disappeared into the woods — ka mauria te //e^ww— carrying the 
bird. Sandy's shooting was over for that time, and he got out of the 
bush with speed." 

WARNING VOICES. 

One of my Ma(»ri friends said : — ** Down near where the *Tararua ' 
was wrecked is the Whare-kaio landing-place. Here the rock has 
kelp on it, and it is the only place along there where you can get kelp 
for pohas or eel bags. Wiien the natives had cut enough kelp a 
voice from the bush would say, * Kati, waihoki ina atatau,^ If you 
didn't stop when you heard tliat voice you would die when you 
reached home. If a Maori did not obey the spirits disaster would 
overtake him." 

" Wlien the Maoris were at Tautuku, and the women went out to 
cut flax," said anotlier old man, ** the Maeroero would warn them 
when they Iiad cut enough, and evil befell them if they did not heed 
the warning. The Maeroero lived down there, hence are the wooded 
hills there called Puke-maei-oero. It was also said of other places 
that spirit voices warned the people when they had caught enough 
fish. These are all wliat the white man would call mere fairy yarns, 
but they were believed in by the people of old." 

Altlumgh no warning voice emanated from the Burning Plain 
near Pomahaka, the wreaths of smoke from the lignite were a warn- 
ing. Tradition says that the Maoris tried to cook on the steam • 
escaping fnmi the ground but the food went black, and thereafter 
the natives shunned the locality. 



Digitized by 



Google 



214 JOURNAL Of THE POLYMESfM SOC/ETY'. 

THE TABPO. 

What the taepo is tYie collector cannot define. The dictioiiarj' says 
it means *^ a goblin, a spectre/' but in tlie south, at least, it seems 
capable of a wider use than this definition would imply. One old 
woman says tliat when she was a girl she saw one. It was at the 
winding part of the Taiari River, known as Te Rua-taniwha, which 
was said to be haunted by the ghost of a karma. The grown-n|>8 
said to the cliildren, ** Don't go there — that is where a taepo dwells," 
• but; the little ones in a spirit of fun went to that spot and shouted, 
" Taepo come out, Taepp come out ! " and suddenly, said the ancient 
dame, they saw a fearsome kind of thing like a shark come out from 
a hole in the bank, and begin throwing water in the air with its tail. 
They waited for no more, but fled screaming from tlie spot." 

An old man naiTates thafc many years ago he went an eeling trip 
from Tuturau up to the Otama and Otama-iti lagoons, and that when 
camped at Whareoka (near Charlt<m) a violent earthquake occurred 
in the middle of the night. He woke up with a start and thought 
the taepo was abroad, or that the eels had come to life again and 
broken loose. Another -tremor came, and when it passed he was 
quite relieved to And he was still on terra firma, whole and sound, 
and that the taepo had not got him. 

TANIWHA AND KARARA. 

The dictionary defines the tamivha as a water- monster, and the 
name has cropped up once or twice. A very old man said : — " There 
is a lagoon on Euapuke Island called Wai-o-tokarire, and in it lived 
a taniwha. It was a wairua (spirit) or taepo and had long hair on it. 
When the people went to get water they could see it floating about. 
They tried to dig an outlet to drain the water, and after some water 
had got away the people could see the cave the taniwha had been 
living in but it had disappeared in the night. The place was tapu for 
a long time after that, and it is still regarded witii awe." 

If taniwha means a water-monster the term karara {ngarara in the 
north) must also mean the same in some cases, although it is also 
applied to a kind of big lizard in the south. From a roomful of 
Natives at Colac Bay the collector could only glean this meagre 
information : — "A kind of shark is called tanvoha. There was a 
karara on the Mataau called Kopuwai, and it captured a girl called 
Kaiamio ; " but they knew of no other karara, nor did they know of 
Maeroero, taepo nor tipiM in Murihiku, but ghosts (atna) were 
j^entiful in many places, they said. 

Kound near Orawia (which is correctly Orauwea) in a field some 
rocks can be seen, and these are said to be the petrified remains of a 
karara. It killed men who were out hunting wekas, and finally chased 
a man named Taiari. He ran zigzag to escape it, and it became 



Digitized by 



Google 



Tmdittons mnd Legends. 215 

jammed between two trees and was killed. Another account says its 
habitat was the west side of Hekeia (Bald Hill). It may be added 
that on the west side of the Waiau Biver at Clifdeu are caves called 
Te-ana-o-te-karai-a. 

A demon fish lived in a hole in the Whawhapo Creek at Kaka 
Point. It could turn itself into an eel, a log or a minnow. It had 
a name which the narrator could not call to memory. A child 
fell into Lake Kaitiria long ago and disappeared. It was at a spot 
near where Mi*s. Aitcheson now lives, and the people reckoned a 
demon had got it, hence they called that place Kai-takata (and this is 
the origin of the name Kaitangata as now given by the pakeha^ 
thought the narrator). 

The North Island is **Tlie Fish of Maui," and some time after the 
Tarawera eruption of 1886, Tare Wetere te Kahu, of Otakou, visited 
the north, and in conversation witli Wahanui, of the King Country, 
said he considered the great fish had become restive and given itself 
a shake with a result that a scale had flown out through Mount 
Tarawera, thus creating the eruption. 

FAIRIES AND ELVES. 

Not all tiie supernatural beings of the olden Maoiis were grim 
monsters or fierce goblins ; the fairies and elves are in a different 
category. The hill near Palmerston South, known as Puketapu, is 
sometimes swathed with fog as picnickers who have airanged to 
ascend it find out to their disappointment. Tlie Maori legend avers 
that the mists only come on Puketapu when the spirits are holding 
high revel on its summit and sides. Their flutes and their musical 
voices gleefully singing and calling to each other can be heard through 
the white curtain they have imposed between man and themselves. 
My Maori infonuant added that Gnr\o\\B pakehas^ as well as inquisitive 
Maoris in the past, had sometimes tried to *' beat the mist " but their 
endeavours to penetrate the veil of mystery shielding the elves had 
always been unsuccessful. One man said the fairy people are fond of 
playing the kind of flute known as koauau, and it is a female spirit 
who plays in the hills near Catlins. He reckoned these elfln 
musicians came in the canoe ^* Takitimu," but other natives considered 
that these spirit people came in a very much earlier canoe. 

A FAIRY WOMAN. 

Kai-he-raki was a witch woman who lived on the'Takitimu Pange 
— not an ugly old witch-hag, but a young and beautiful witch whose 
comeliness defied Time. She was (apu. A man out hunting wekas 
caught her, but the narrator forgot his name. The man said to his 
captive, ** 2'aku taahine pai" and she answered, ** 2'aku tanepai" He 
got his katiati out to make a fire to ta whakamoe or remove the tapu 



Digitized by 



Google 



216 JOURNAL OF THE POLYMEBIMM SOCIETY. 

from her. He told her to put her f«H>t on the kattati while he worked 
tlie karimaHma on it. Soon smoke came and a little flame kindled 
but she threw bloo<l «>n it, putting it out. She fled but the man 
overtook her and coaxed her bnck. He wiped i\\e kaiwti oavehiWy 
and started again on liis task, but the fiiiry woman repeated her 
previous perfornmnce and this time she escaped for good. That was 
the onl}' man who ever caught her. She was seen afterwards high 
up on the mountains and finally she vanished. 

Another account ends: — "The Takitimu Mountains are still 
haunted. Kai-he-raki has been seen there in quite modern days." 

TOHUKA (wizards). 

Merehau, & Johuka, who resided usually in the Port Molyneiix 
district was living when the white settlers came. He was a magician, 
said one of the collector's informants, and if offended could upset 
canoes which were out at sea^ and lie could do other magic. He had 
a garden between 0-marama and Te Karoro, and he called it Te-au- 
o-Hatane, which means "the gall of Satan," according to the 
narrator, who added, " He was not afraid of Satan as he always had 
his own taepo with him," and continued, " The powers of the tohukas 
were wonderful. Matamata, a priest or prophet of the old times, was 
appealed to in storms at sea. The tohuka would use a firestick, and 
say karakia, and two whales would come alongside the canoe and keep 
it from capsizing. The tohuka would give each whale a hair of his 
liead. Rakitauneke was a famous tohuka of old, and had a guardian 
whale Tu-te-raki-hua-noa, and also sometimes one (;alled Matamata. 
One day the former of these whales appeared off Moeraki, and tlie 
children cursed it, and its owner in anger sent a tidal wave which 
drowned them. The creek they were standing by had been fresh water 
till then but it has been brackish ever since. Its name is Ka-wa." 

There is a place at Stewart Island which, the collector was told 
years ago, was called Hingaringa becan.se a Maori had lived there 
who suffered from leprosy of the hands, but a well-informed kattmatua 
(or elder) says it is named from a famous old wizard, Rikarika, who 
lived there long ago. 

" Te Maraki was a tohuka, and was a cousin of Karaki who was 
the father of Matiaha Tiramorehu. When Karaki died at Moeraki 
the people did not observe the proper burial custtmis, but went on 
working. This offended Te Maraki and he brought whales and evil 
fairy fish roaring on to the land, where later they died. The people 
were soiTy for their actions, and in answer to their entreaties Te 
Maraki sent the great fish hack to sea alive. Since then the water, 
which before that had been fresh, has been salt in the Moei*aki 
creeks." (For comparison with this narration see "Journal Poly- 
nesian Society," Vol. XXVII., page 96.) 



Digitized by 



Google 



Tradi'Hons and LBgendt. 217 

One of the old men was inclined to rank Bakaihautu as a magician 
or as an undoubtedly clever man. '^Rakailiauta," he said, ** was the 
first of the Waitaha tribe, and he went through the South Island 
planting some of his tribe at Oreti, some at Molyneux and so on. 
When he came along and a plain was dry, and he felt thirsty, lie 
stuck in his ho (spade) and lo! he had a pool to drink from." 

A DREAM AND A LIZARD. 

A southern Maori who visited the Hokanui Hills on fonr annual 
toekaAwinimg trips (1867-70) tells the following story: — ** His brother 
and he were camped <m the Otamatea stream, and one night he dreamt 
he was fighting a big strong man (the narrator is himself a tall man) 
and that he finall}' tlirew a big stone at this man and killed him. He 
woke with a start, and his brother asked what was the matter, and lie 
told his dream, and said he would know tlie place he dreamt alxmt if 
he saw it. Next day as tliey were journeying he saw the place. The 
white men had built a sheep-dip near it. They looked to find the 
stone of the dream, and not seeing it they dug down and found two 
big stones one on top of the other, and charcoal and burns underneath. 
They were astonished and said karakia, which the narrator would not 
repeat or it would lose its mdna. Later that day they climbed a }>eak 
with a new trig station on top. Near the sunmiit was a rock as big 
as a cottage, and in a crack facing the N.W. was the biggest lizard 
he ever saw. It was a kavara probably two feet long. He picked up 
a big stone and hit it, and it jum2)ed over the rock. Going round tlie 
brothers found it lying on its back, belly up, and tliey killed it. They 
lit a fire of scrub, and burning it left, in case otiier lizards might 
follow them. They came through the Waimumu Gorge, and reached 
Tuturau with as many tcekan as their horses ctmld carry. Old Karaka^ 
a brother of Mailiaroa and a man of mUntt, was there and he told 
them that gorge was a place of worship where the old tohukaa used to 
go. He said it was a good thing they had killed the lizard and so 
averted the evil of tlie dream." Tlie narrator considered that the 
charred place where the twq stones were found had been an ahi-iapu 
(sacred fire) in connection with the rites of travellers going that route. 
A part of that ceremony consisted in ligliting a fire and burning a 
hair of the head. 

In the old days, continued tlie narrator, lizards were kept as pets. 
One such was found at Motu-kai-puliuka, a clump of bush east of 
Kaitiria (now called Lake Kaitangata.) It was named Te-lioro-mokai 
but it got away after some time, and was last seen in the creek Te- 
wai-a-kiri near there. A small ridge there is named after this 
celebrated pet. Tlie narrator further said that some of the old Maoris 
ate lizards but he had no particulars of it. (See ** Trans. N.Z. Inst.'^ 
Vo]. VII., p. 295.) 



Digitized by 



Google 



218 JOURNAL OF THE POLYHESiAH SOCiETY. 

THE TIPUA AND PUKUTUOBA. 

** Up at Waiiaka," said one of the old men, " is an island called Te 
Pae-karara. There is also another island called Taki-karara after a 
man who had a fishing station at the lake which was called Taumaiiu- 
o-Taki-karara. He stood on a clnmp of vegetation on a point of land, 
and one day tlie point floated away with a noise like a hird. 
Unknown to him there was a tipua under it, and it is said to still drift 
about. Tliis is probiibly what started the stoi-y about the floating 
island. Taki-karara left the distnct as it was too uncaimy for him." 

IHptia is translated as a " goblin, ogre, monster, demon, faiiy, 
spirit," so there is a wide choice of meanings. Mr; D. Monro, writing 
in 1844, says," ^^ a floating island is said to sail about on one of the 
lakes at the source of the Molyneux." Huioihuini's map, drawn for 
Sliortland, in 1844, says of Lake Hawea, ** Here is a floating island 
shifting its position with the wind," and of a place on Hawea's shores, 
"Turahuka — the abode of a tipuay The question now remains — 
althougli an island in Wanaka is named Taki-karara, should not the 
floating island Taumanu-o-Taki-karara be located in Hawea, in 
accordance witli Huruhuru's statement ? The collector will endeavom* 
to ascertain this point later on. 

Mr. Monro also writes : — " Tlie puhatuoh is another wonderful 
animal of the southward, told of by the old men. Under a different 
name he is heard of in the north. A gigantic animal of the lizard 
species, most dangerous to humanity." The infonnation the collector 
received was that the pukuUima was a kind of aquatic monster, and 
that one of them haunted Lake Wakatipu. liawiri te Awha who 
guided a white party to the lake in January, 1859, was very much 
afraid to venture near the edge of the lake after dark lest the demon, 
which he called pukuluola, should seize him to his destiniction.* 

RAKAU-TIPUA. 
The collector again has recourse to the interesting notes of Mr. 
James Cowan for the following item : — " At Te Muka stood the tree 
called * Hine-paak a ' — a kahikatea tree. It was he rakau-m&na-Ufpu 
(a SHcred tree) and was fenced round. Of this tree strange tales ai*e 
told. A pakeha chopped it down but on returning to cut it up he 
found it standing erect again. Its god had raised it. No one could 
fell it ; no one could buiii it. After all the bush around it had been 
felled and buiiit it still stood unharmed." 

* The Puku-tuora is, perliap8, identical with the Tuoro of the uorth. The 
Maoris describe this mythical auimal (or fiiih) as being like a very large eel witli a 
great lump in its tail. It was said to be eight or ten feet long, and as thick as a 
man's body. It gave chase to any one approaching the lagoons in which it lived, 
and the only way the Maoris had of escaping it was to cross land off which the 
fern had been burnt. — Editob. 



Digitized by 



Google 



Tradithnn and Legends. 219 

The collector hits heai'd ramors of bewitched ti'ees which could move 
iroui place to place. So far lie has no particulai*s of such, but his notes 
mention one or two trees which, although not in the niiraculaus 
category, may be nientione<l here. One old man said, "There is a 
toiara tree at the mouth of the Pou-mahaka (Pomahaka) River named 
Rfiki-ihia be.-mnse that great chief camped under it and caught a lot 
of eels there. It is on the south side of the Pomahaka where it joins 
the Molyneuz. A big toiara tree grew close to a rocky face near the 
Mataura River below East Gore. Its name was Ota-karamu, but how 
it got that name I do not know. The hills ix>nnd there are now called 
by that name. The pakeha cut that tree down." Anotlier old man 
said : — " The tree Rata cut down for a canoe was a totara. This tree 
does not grow in Katmka islands, so what land would it be ? It is the 
chief of the forest trees and must be very ancient as it was the son of 
Muniuhako in the days of the gods."* 

MAORIS VERSUS SEALERS. 

The conflicts of the white sealers with the native inhabitants form 
an exciting chapter of early southern history, and were mentioned by 
some of the old men. One related these incidents: — "Up about 
Arawhata or Okahu (Jackson's Bay) on the West Coast was a kaika 
(village) with perhaps 200 or 300 inhabitants. The settlers were then 
at Arnett's Point further north than the kaika, and one night some 
canoes went up to the sealing station on a raid to try and acquire some 
of the white man's treasures. The Europeans were alert, however, 
and fired on them, but without killing anyone. In the scuffle tlie only 
man killed was a white man, one Perkins by name. The Maoris 
returned to their home and some time after the sealers, among whom 
was tlie famous Chaseland, pulled in to the shore fronting the kaika^ 
and in revenge shot some of the natives from the boats. They then 
landed and slaughtered all who did not escape into the bush. When 
Chaseland was roused he became a frenzied fiend. Among his other 
acts he seized a child, Raminkiri, whose father and mother had been 
killed, and dashed her head on a rock and left her for dead. After the 
•sealers had done all the miscliief they could they left and the surviving 
natives crept out of the bush and returned to tlieir desolate h<mies. 
They found the little girl living and revived her and she died at Colac, 
an old woman, some fifteen or sixteen years ago. Often in past years 
did I hear her tell Chaseland what a savage brute he was when he was 

* This refers to the story of the building of the (celebrated cauoe called 
Maiiu-ka-rere, which was hewed out from a tree in the forest of wbioh Rata-i-te- 
wao was the guardian. The Rarotougans say the tree was a Maota-mea (which 
grows in Samoa), and that the tree was growing in the island of Kuporu, i.e., 
Upolu of Samoa. See this •Journal,'* Vol. XXVIII., p. 140, par. 270. This 
shows how incidents occurring in other countries become localized. — Editob. 



Digitized by 



Google 



220 JOURNAL OF THE POLYMESiAM SOCiETY, 

young, and she would rebuke him for his part in the nifissacre. Chnse- 
land did not like it but he had nothing to say in self-defence. But tliis 
is not the end of the story of the cruel deeds done them. The sealers 
pulled round to Milford Sound and there met natives from the south. 
The Maori name of Anita Bay is Tauraka-o-Hupokeka because the 
old chief named Hu-pokeka used to come round there for gi*eenstone. 
These people were ignorant of the row between the sealers and the 
Westland natives, and they had done the sealers no harm, yet they 
were killed without mercy. The victims had only Maon weapons, and 
were shot down like rabbits.' Hu-poheka was shot while standing on 
a rock in Anita Bay, and such of Ids people as were still living were 
placed in a canoe and towed by the sealers round to Whareko, the next 
bay south of Milford ; and there to finish the sport the canoe with its 
helpless crew was let go in the surf. The breakers dashed it to pieces, 
and not a soul was left living/' 

FURTHER ATROCITIES. 

" At Paringa, on the West Coast,'' said anotlier of the old men, 
" the sealers killed some seals, and wanting to go after others the}' 
told the Maoris who were there to skin the seals and to do what they 
liked with the carcases. The sealers having said this departed after 
more seals. Unfortunately the language they had used was not plain 
and the natives did not properly understand it, so tlie Maoris made a 
big fire and singed the hair off the seals as they usually did. When 
the sealers returned they were veiy angry at what had been done, and 
they pulled their boat a little way out from the sliore and fired at the 
people, killing one and breaking the aim of a chief named Kahaki. 
They then made off. The Maoris walked overland to Otago Heads 
where they considered the matter- A party left for Stewart Island to 
obtain revenge on s(mie sealing gang or other, and eventually they 
caught and killed some white men at Murderers' Cove on South Cape 
Island. This is maybe, where the girl threw the cape over Jimmy the 
Boy. Waliia, a chief from Otago Heads, stopped the killing by say- 
ing they had had enough revenge and to figlit no more. The 
Europeans playing tricks on the native women made trouble some- 
times." 

Another aged man said in passing": — " At Murderers ' Cove on 
Taukiepa (South Cape Island) a massacre took place. Only two, a 
woman and lier child, were saved and she hid when tlie crew were 
being killed. Slie was either a European or Kanaka woman, and the 
child and she wei*e rescued by a later boat of sealers. As regards 
Caddell, or Jimm}*^ the Boy, I think he acted as interpreter when peace 
was made at Bakituma (Preservation Inlet) between the Europeans 
and Maoris, and then he went over to Sydney with his own race." 



Digitized by 



Google 



Tradiiions and Legends. 221 

The late Hon. Tame Paraia said to the collector : — " At Wliaie- 
akeake (Murdering Beaeli) the Maoris and sealers had a fracas. 
Matahaei^e (the father of Riuiurapa) although a small man seized a 
sealer and began carrying him, and calling on his people to come and 
kill tlie man. While they were struggling the boat escaped, leaving a 
boy Jimmy, and a Kanaka from Calcutta named Te Ann as captives. 
The only pakehu killed was the one on the back of Matahaere. The 
two prisoners were taken to Hakiura (Stewart Island). A girl threw 
a mat over the boy to save him, but I do not know her name. Jimmy 
was given a wife named Pi, but they had no children. Te Anu maiTied 
a woman and had a son called George Turi, but no descendants are 
now living. Te Aim lived at Colac Bay where I can remember him. 
Jimmy went over to Sydney in the end and never returned. In regard 
to tiie killing in 1817, it was at Hobartown Beach, and was another 
affair altogether." 

SWIMMING. 

Th6 question of acquatic sports cropped up in conversation with the 
old men, and here is what they said : — ** When a Maori swam, with 
his shoulders out, we called it, * He kau tn,'* Sometimes the young people 
would assemble on tlie bank and one would call out * ka rukn tauay 
and they would all dive together." 

" There were two kinds of swimming our old people did as far as I 
know. One was with the body uprig})t and working the legs, and the 
other was on the side with only the side of the head showing. At 
Ruapuke there were no surf beaches. The people swam in fresh- 
water lagoons ; the men and women bathed together." 

" Swimming was called kmi. The younger people would bathe in 
the sea ; the older ones preferred the wanner water of the lagoons. 
At Ruapuke there was no^bathing in the oj^en sea — the people bathed 
in the Tau-o-te-niaku lagoon but not in the one called Wai-o-to- 
karire." (It was haunted — see supra.) 

" The Rapuwai people when swimming lay on their bellies with 
their elbows close to their sides, and hit the water with their hands — 
he^ice their name. The old Maori way was to swim like a walk in the 
water with the water up to the armpits. You worked with your legs 
and elbows, and it was surprising how fast one could go. This style 
was called kau-po^ 

** When I was a boy I saw three kinds of Maori swimming — kau- 
tu, swimming upright; kau'tahi, swimming lying on the side; and 
kau-tuara, swimming lying on the back." 

SURF -BATHING. 

At least four of the old men mentioned the sport of surfing, as 
follows: — "The young Maoris would swim out with a short board, put 



Digitized by 



Google 



i%% JOURMAL OF THE POLYMeSiAM SOCIETY. 

it under the chest and shoot in on the waves. I remember round at 
Kukararua (Hunt's Beach, Westland) we were at it, and a white man 
named Baker would tiy it. He was a big, heavy man, and when he 
came in his board struck the shore and almost stunned him. His chest 
was rather severely hurt." 

"The board used in sui*fing was called a papa, and it requires 
certain practice to use it. You must keep the end of it up just as you 
reach the beach or it will dig into the sand and perhaps break your 
ribs. The board was about four feet long perhaps, and came in like 
an arrow. I was round at the West Coast diggings, and the beaches 
there are very suitable for it. Another sport was when the boys would 
take a taivai (a kind of canoe) out and come in through the surf. 
Tiiey would capsize sometimes but that was all the more fun." 

" I never saw tlie sport of surfing, but know that a papa or surf- 
board was used. I have heard that in the whaling days old Takata- 
huruhuini went sui'fing in the bay at Port Molyneux. He was a 
descendant of the people who came south in the Makawhiu canoe." 

The late Tare te Maiharoa said : — " Take kelp off the rocks and 
diy it as for pohas or kelp bags [to presei-ve birds in]. Take two of 
these bags and tie them together about two feet apart. Blow them 
up, and having got them out beyond the surf, put one on each side of 
you from the armpits to the hips, lie on the flax connecting them, and 
come in with the breakers. It is fine sport and you cannot drown. 
This was an old pastime at Moeraki, Waikouaiti, and other good 
beaches, and was called para, (He pronounced it pala.) In the old 
Maori days there were very few sharks about — they have only come in 
any numbei*s since the European fishermen throw the fish -heads back 
into the sea." 

The names papa and para are interesting. The collector looked up 
Tregear's Dictionary, and in it he notes that in Hawaii a sui-f-board 
is called /)fi/>a, and in Tahiti it is named papahoro. As for para the 
nearest appropriate meaning seems to be " the half of a tree which has 
been split down the middle " (and hence may be cut down into a surf- 
board) but perhaps Maori scholars could help to explain the term para, 

TATTOOING. 

Wishing to know about tattooing in the south, the collector asked 
about it and received these replies : — " The usual name for tattooing 
was mol'Oy and the work of tattooing was called * ta ki te nwko,^ Some 
of the old people, I i*emember, were tattooed, but no one has been for 
many yeai^s now." 

" The old man who brought me up was tattooed on one side of his 
face only — a thing we called kane or katce* Old Mrs. Paina at Colac, 

♦ Knuae.— Editor. 



Digitized by 



Google 



was tattooed with two straight lines across her face with dots in the 
middle, but I do not know the name of that kind of tattooing." * 

'* Tattooing in straight lines is a very old way and is called ttihu 
We call writing nowadays iuhtiuhi. The usual name of tattooing was 
moko, but the different tribes had different styles. That done on one 
side of the face from nose to foi*ehead was called titvhanay a pattern 
scroll all over the face was called huriitta, while tattooing on the hips 
was repe^ 

^^ My father was tattooed in a single line across one cheek from ear 
to nose« This was called tuhi^ altliough the ordinary face tattoo was 
called moko. He was tattooed on the arms also, and this we called 
tiatia. Old Koraka was tattooed on one side of his face only, and this 
made him look fierce. His body was tattooed in the tiatia fasliion 
down to the waist, so that when he cast off his kahahu (cloak) and 
possibly his maro (waist-doth) also — when fighting — the tattooing 
would show. The word tiatia really means 'pierced,' but it was 
applied to that tattooing." 

JOTTINGS. 

. Tu-mataueka. In Mr. 8. Percy Smith's ** Maori History and 
Traditions of the Tai*anaki Coast," at page 537, we read of the murder 
at Kapiti of tlie southern chief Tu-mataueka. When news of this 
reached Buapuke a descent was made on Tutaeka-wetoweto (Lord's 
Kiver, Stewart Island) where two North Island Maoris were killed in 
retaliation. My informant said, ** There is a beach at the mouth of 
Lord's River called Ka-one-o-Whitiora, after a Thames native who 
was killed there in revenge for the murder of Tu-mataueka, a brother 
of Haereroa Toheti, at Kapiti Island. Whitiora was with a pakeha 
called * Scotch John,' who was building a boat. His brother, Ueka- 
nuka, was also killed, but another Maori escaped. Their father 
afterwards came down to Otakou and got some greenstone as utu '^ 
(or payment). 

Turihuka. In the *' Memoirs of the Polynesian Society," Vol. IV. ^ 
page 242, we read that Turihuka, the wife of Tamatea, died (about 
1350 A.D.) on a high ridge in the South Island (presimiably in the 
Waitaki region from the narrative). In retailing some of the 
nomenclature of the lakes districts, an aged southerner said, '^ A big 
hill wliich the Europeans call * Old Woman's Hill,' was called Turi- 
huka after a woman who died there long ago." He was referring 
to tlie mountain which the surveyors named Breast Hill. It is 5,146 
feet high, and lies close to Lake Hawea on the eastern side. This, to 
the collector's mind, identifies it as the place where the wife of 
Tamatea died. That Tamatea visi4;ed there is quite probable as he 

t Probably mako'kuri. — Editob. 



Digitized by 



Google 



224 JOURNAL OF THE POLYMESIAfl SOCIETY. 

was evidently a great explorer and traveller, for his name is associated 
ill tradition with Te Rua-o-te-moko, Takitiinii Mountains, Dome 
Mountains, Waimea Plains, West Coast Sounds, etc., so it is likely 
he visited the lakes and also the Waitaki River. 

Te Rua-o-te-moko is the Maori name of the densely wooded and 
mountainous country between the Waiau River and the Fiords. It 
was the retreat of broken tribes but, it may be noted in passing, these 
people are described as men, not as elves or phantoms as would 
probably be the case were the traditions ancient. The name of this 
vast tract of rugged land is sometimes g^ven as Te Rua-o-te-moa but 
the former name is more probably correct, as it is said to have been 
named in commemoration of Tamatea having pits {ma) dug to get 
pigment for tattooing {moho) when he explored the Waiau country. 

Te Au-kukume was the name of a ^flfi^o south of theTaieri mouth; 
it was a fishing camp of Te Raki in modern days. This is a name of 
some significance in Kati-Mamoe circles for Te Au-kukume was the 
wife of Hotu-Mamoe, after whom the tribe was named. 

Timaka. ** At Tautuku Beach there is an old burial ground," 
wrote the late Mr. W. H. 8. Roberts, **in which are several Maori 
graves. At the head of one is a slab of Australian cedar, with the 
inscription : * Sacred to the memory of Temnc who departed this life 
September 25th, 1846.? " One of my informants said, "Timaka was 
a woman who died at Tautuku and was buried in the whalers' 
cemetery there. Her mother, Kiwi, had died near Kaitangata and 
was buried in a hapua (lagoon) there known as Te Karohe. A man 
was fishing in that lagoon, and poking a stick under an overhanging 
bank her skeleton came up, and it was then buried in a landslip near 
Stirling. A boy named Temu died about the same time. He was out 
with an eeling and birding paHy, and going out alone failed to return. 
Search was made, and his body was found stone dead and was taken 
down to Port Molyneux and buiied. The cause of death was 
unknown. This sudden fatality occurred near the falls on the Kai- 
hiku River. The word * Temuc ' on the slab should be Timaka." 

Haki-te-kura. The fact that this celebrated lady swam across 
Lake Wakatipu was mentioned in these articles (Vol. XXVT., page 
83) but one of the old men has given further details. She swam from 
about where Queenstown now is, and apparently she must have set 
out in darkness, for she steered her course by Cecil and Walter Peaks 
whose tops in the dawning light she watched twinkling and winking 
at her like two eyes, hence their name Ka-kamu-a-Haki-te-kura. 
She landed on Refuge Point and lit a fire, and that is why it is known 
to the Maoris as Te-ahi-a-Haki-te-kura. 

The voice. Either the voices mentioned were exceptionally sten- 
torian, the hearing of> the Maoris extremely keen, or the localities 
possess great acoustic properties if the following bona fide narration 



Digitized by 



Google 



Traditions and Legends. 225 

by an aged Maori is accurate. '' Tlie celebrated Kati-Mamoe chief 
Baki-ihia had a voice that could be heard from Wharepa light over 
to the hills east of Kaitiria (Lake Kaitaiigata). In case you scarcely 
credit this, I may mention that my brother's voice was once heard at 
a distance of eight miles. He was standing on the hill Uhi-whitau 
. (near Kaitangata) and his voice carried to Akatorea. Raki-ihia had 
a very powerful voice, and he could shout at Wharepa and be heard 
at Uhi-whitau." 

Obituary. It is with i*egret that the collector notes the narrowing 
of the circle of his aged native friends. Hone te Faina, Ratimira te 
Au, Wiremu te Awha, of Oolac Bay, and Tuhituhi te Mai*ama, of 
Bluff, have all gone during the last year or two. Tuhituhi was a 
brother of John I'opi te Patuki, and the newspapers gave his age as 
110, although the collector figured it out as about 90. He and the 
late Mrs. Gilroy (died at Bluff, aged 86) gave much valuable infer- 
mat ion about ancient Maori place-names. I'here died the other day 
at Puket^raki, Bia Tikini, aged 105, she being seventeen years old in 
18S1, when Te Rauparaha captured Kaiapohia. The collector called 
on her in 1915, but found her very deaf. She was tattooed in the 
tuhi style, each side of her face being adorned with two straight lines 
from temple to mouth and from mouth to ear. Tare te Maiharoa has 
passed away in South Canterbury in his 70th year. He gave much 
information that has appeared in these articles, and was a great 
stickler for accuracy, he was always most anxious that the correct 
history should be preserved. The collector always found him a mine 
of information and was looking forward to further interviews, but a 
fall from a stack cut short a life that, as far as health and activity 
went, seemed destined to continue for years. He was the greatest 
authority left on the Waitaha lore, and his death leaves an irreparable 
blank. 

(To be continued, i 



Digitized by 



Google 



NIUE FOLK-LORE. 



By G. N. M0RBI8. 



THE STORY OF THB YEKA AUD THE KALE. 

THE home of a Kale was in a cave which is known as Anakale, 
and a Veka lived at Tukaiavi nearby. The Kale went to visit 
the Veka but on arriving at Tukaiavi there was a quarrel between 
the two. The Kale said, ^* The long-leaved banana and the sugar- 
cane are for me alone. For you there is plenty of filth where the flies 
gather in the open spaces." The Veka was angry at the insulting 
words, and he sought a plan to kill the Kale. The Veka went down 
to the reef to a place where a Clam lived, some twelve feet from the 
foot of the high cliff at Tukaiavi. The Veka put his feet near the 
Clam-shell, and there danced and sang to entice the Kale to come 
down from the top of the cliff where he was sitting. This was the 
Veka's song : — 

Come, come Kale, come, come Kale, 

Here is a fine place for your feet. 

Come here and be tickled, 

come and be tickled ! 

So the Kale came down to the Veka and the Clam, and put his 
feet into the Clam-shell which closed upon them and held them fast. 
Then the Kale cried and screamed to the Veka for help, struggling 
this way and that to free himself but unable to do so. The Veka ran 
up the cliff and danced in his gladness. He sang a second song in 
mockery : — 

Kale, Kale, 

Now chew your sugar-cane, 

Now eat your bananas, 

Tou tiled to fool me into eating filth, 

filthy Kale, O filthy Kale! 

But afterwards the Veka began to be sorry for the Kale, and he 
sang another song to help him to free his feet : — 
Rise quickly, Tide, 
Rise quickly, Tide ! 



Digitized by 



Google 



M/uB FM^Jore. . 227 

And the tide came in and the moutli of the 01am opened so that the 
ieet of the Kale were freed. He eame up with an incoming wave 
which the Yeka did not see, but he was surprised to see the Kale on 
the top of the cliff. The Kale caught hold of the Yeka and struck 
him on the head with a branch of the Toi tree. The Yeka begged 
for forgiveness, calling the Kale by his honorific name, Manatafeiki, 
but the Kale would not forgive the Yeka and continued to strike him, 
splitting his head in several places so that the marks of the strokes 
and the blood can be seen to this day. Afterwai-ds they made another 
agreement about the food. The Yeka agreed to let the Kale have 
the bananas and the sugar-cane, but the open places, where the flies 
live, were to be for himself. That is to say filth was to be the Yeka's 
portion. 

Notes. — The Yeka (Maoii, Weka) is long extinct in Nine having 
been exterminated by wild cats and dogs. The bii-d was evidently 
not particularly dainty as to its food. 

The Kale. This bird is still plentiful and is similar to the New 
Zealand Pukeko. It plays havoc with the sugar-cane and banana 
plantations, and is the only bird not protected by law in Nine. 

Tukaiavi is the place where the Government Jetty at Alofi now 
stands. The cliff is just to the north of the jetty. 

The native version was written by Harry Lupo, Government 
Interpreter, from information obtained from Togiafulu, Ilea and 
others. 



KO E TALA KE HE KALE MO E VEKA. 

KO e kaina ne nofo ai e Kale, ko e ana, ti ui ai e ana ia, ko 
Anakale. Ko Tukaiavi e kaina he Yeka, ti fano e Kale ke 
feleveia mo e Yeka. Kua hoko atu a Kale, ke he kaina a Yeka, ko 
Tukaiavi, ne taufetoko e Kale mo e Yeka, in pehe e Kale : *^ Ka 
lauleleva mo e lau malikalika, haaku a ia ; ka momolago he mala- 
malaega, haau ia." Ti ita e Yeka he kupu kelea e Kale mo e kumi 
lagatau ke mate ai e Kale. Ti hifo e Yeka ke he uluulu i tahi ke he 
mena ne tu ai e Gege ; hogofulu ma ua e futu he vaha mamao mo e 
pokoahu, e mena ne tu ai e Gege ; ko Tukaiavi e higoa he maga tahi 
ia. Ti uta e ia e tau hui haana mo e fakaeleele he gutu Gege i fafo 
mo e koli ; ti uhu ai e Yeka a lologo ke fakaohoohoaki e Kale ne nofo 
he feutu ke hifo age. Pehe e lologo : — 

'' Kale, Kale, o hau ke 

*^ tamai nove fakatu 

** he mena malie e 

** nukua maineine noa 

** kua maineine noa." 



Digitized by 



Google 



228 JOURMAL OF THE POLYMESIAtl SOCIETY. 

Ti hifo e Kale ke he niena ne uofo ai e.Yeka nio e Gege, ti faka- 
olo e ia e hui ke loto he Gege ti kati ai e gutu he Gege ti apitia ai 
haana hui. Ti ui atu e ia ke he Veka ke haii ke lagomatai a ia hakua 
tagi fakatutu a ia, kapa ki he, kapa ki ko, ai maeke. Fakakopa a 
Yeka hake nolo he feutu mo e koli he fiafia ti uhu foki ni he Veka 
taha lologo f akafiu. Pehe loiogo : — 

'* Kale, Kale, 5 gaii ho to 

'' kai ho futi fakavaiaki 

** ti vala te. Tetekale, tetekale." 

Kua mole ia ti tupu ai e fakaalofa a Veka kia Kale, ti uhu ai e ia 
e lologo ke maeke ai e hui he Kale. Pehe lologo : — 
** Tahi hokohoko vave. 
" Tahi hokohoko vave." 

Kua hoko e tahi ti mafa e gutu he Gege, maeke agatuha e hui he 
Kale. Ti hake a ia he peau, ai kitia he Veka, ka e ofo ue Veka kua 
hoko hake tuai e Kale. Ti tapaki e ia e Veka, tamai e la Toi fahi 
aki e ulu, ka e ole atu e Yeka kia Kale pehe : '* Manatafeiki, Mana> 
tafeiki " (ko e taha higoa lilifu ia he Kale he vaha tuai). Ka e fahi 
ni he Kale e Yeka ti maihiilii loga Iiaaiia ulu ; ko e mena ia ne 
avaava ai haana ulu mo e kula he toto ke hoko mai ke he vaha nai. 
Mole ia ti liu taute ai foki e laua e pulega ke he tau mena kai. Pehe 
a Yeka : Ka lauleleva mo e lau malika, ha Kale a ia, ko e futi a ia 
mo e to ; ka moniolago he malamalaega, ha Yeka a in, ko e te a ia. 

Ko e mena ne tu ai e Gege, ha he mena kua toka ai e uaafu ai 
nai, ke he faahi tokelau. 



Note. — The Hindus have a very similar story to this, but I cannot place my 
hand on the reference.— Editob. 



Digitized by 



Google 



THE MAORI BELIEF IN THE SUPERNATURAL 
POWERS OF CERTAIN AXES. 



rilHERE are many references in past numbers of this ** Journal " 
JL to some of the celebrated stone axes of former times, which 
were supposed to be endowed with supernatural powers ; they were 
in fact looked on as gods. These notes are intended to show that a 
very similar belief was in very ancient times comtoon to other races. 
But first it may be as well to recall the traditional history of the two 
most famous of these Maori axes. 

These two axes were named **Te Awhio-rangi" and **Whiro-nui." 
An attempt at the translation of these names may be made, and it 
will be seen that they both embrace the same general idea. The first 
name means " The heaven -encirder," the second "The great twirler." 
And when we learn the purpose for which, according to tradition, 
they were first used, the meanings seem somewhat appropriate. The 
axes are first mentioned in Maori tradition in connection with the 
earliest dawn of their history and mythology. After Bangi-nui (the 
Sky-father) and Papa-tua-nuku (the Earth-mother) were separated 
by their offspring tlie Whanau-rangi, the minor gods, it was found 
that they clung together in lingering embrace which prevented the 
full enjoyment of the light of the world by the gods, and to a certain 
extent nullified the object of the separation of their parents, which 
was to allow of their offspring proceeding forth fi*om the aeons of 
darkness represented by the close embrace of their parents, to the 
Whai-ao and Te Ao-marama — the world of being and of light. These 
celebrated axes were then used to sever the aims of the two parents, 
and cut the '* props" which were erected — principally by the god 
Tane — to support the Sky-father in the position he now occupies as 
represented by the dome of the sky. 

According to Maori belief these same axes were brought away 
from the Father-land — the site of creation — and accompanied the 
migrations in their wanderings in the Pacific to the time when one of 
the migi'ations that had charge of them settled down in Tahiti at a 
date which is somewhat uncertain, but probably about the tenth 
century of our era. 

The next we hear of the axes is when the '* Takitimu" canoe, on 
her voyage from Tahiti to New Zealand about the year 1 350, met 
with a gale of wind at that part of the ocean called ''Tuahiwi-nui-o- 
Hine-moana " (The-great-ridge-of-Lady-ocean), when the same axes 
were used by the priests to ** slay " the gale. In our " Memoirs," 
Yol. lY., p. 224, is this statement : ''When the canoe arrived at this 



Digitized by 



Google 



230 JOURMAL OF THE POLYMESiAM SOCiETY. 

place the priests took the axes * Te Awhio-rnngi ' and * Whiro-iiui ' 
from the depository in the stern of the canoe, where they were kept 
in a calabash named * Ahuahu-te-rangi.' These two axes were very 
tapu^ and were used in the poipoi (or * waving ceremony ') offered to 
the gods Kahu-kura (and others). Tupai (the priest) took hold of 
* Whiro-nui ' axe, and tlie following is the karakia (invocation) used 
by tliose priests to ' fell ' the easterly seas of Tahiti." Then follows 
the long invocation in which the names of the axes are mentioned 
more than once. After this the narrative goes on (p. 226) : ** Then 
were the two axes used to chop the waters by Te Itongo-patahi and 
Tupai (priests) ; and thus were the seas severed and spread abroad. . ."" 
Herein we observe the supernatural powers accredited by the Maoris 
to these celebrated axes. 

When the migration above referred to visited the Whanganui 
Biver, they met people of one of the other migrations, and a marriage 
was arranged between a chief and chieftainess of the two parties, in 
which the ** Awhio-rangi " axe was given to the latter's tribe as a 
marriage gift (see p. 244, he, cit.). From that time this axe remained 
as a most valuable property with the Nga-Rauru tribe of Wai-totara. 
Seven generations ago, during some tribal stress probably, the then 
custodian, Bangi-taupea, hid the axe away, and it was not discovered 
until the year 1887, when some great ceremonies accompanied its 
exhibition to the-tribe, as described in this ** Journal," Vol. IX., p. 
229. The axe is now preserved with the utmost care by the ti-ibe, 
and is hidden away so that only its guardians know where it is. It 
is so sacred that no white man has ever been allowed to see it, though 
a desciiption has been obtained, from which it would appear to be 
one of the large axes made by the Polynesians, all over the Pacific, 
from the giant Tiydacna shell. 

Of course the story of these axes having been used to sever the 
limbs of the Sky-father and Earth-mother, has been applied to them 
to give the axes additional mdna^ or power, prestige, etc. But there 
can be little doubt one, at least, was brought herein the *'Taki-timu " 
canoe in the fourteenth century. About ** Whiro-nui " there is some 
doubt. One account says it was left in Tahiti. 

It is tolerable clear from the above that these axes were endowed 
in the belief of their owners witli supernatural powers by which 
miracles were worked, that in fact they were gods. One of our most 
learned Maoris says of these axes that they were possessed of md9M- 
atua, that is, god-like-powers. 

That other races had similar beliefs the following quotation will 
show. We all know of the most interesting and valuable discoveines 
made duiing the last twenty-five years in the island of Crete, in the 
Mediterranean, during whicii the ancient Minoan civilization was first 
brought to light. In a book by Prof. R. M. Burrows published a few 



Digitized by 



Google 



The Maori Belief in the Supernatural Powers of certain Axes, 231 

years since entitled *• Tlie Discoveries in Crete,** p. 113, he remarks 
on the sign of tlie double axe form on many of tlie building stones in 
the ancient ruins of Kiiossos, and says, ** The evidence is overwhelm- 
ing from every site in, Crete that the double axe, like the sacrificial 
* horns of consecration ' with which it is often found, was intimately 
connected with religious worship ; and it is highly probable that, like 
the pillar, and less commonly tlie shield, it was originally regarded 
as the visible habitation of the divine spirit. , . . its appearance 
finally among the cult-objects in the sacrificial procession on the 
sarcophagus from Hagia Iriada (south coast of Crete) is evidence 
enough and to spare." 

" Starting as a kind of fetish in early Aniconic days, the axe 
survived as an object of worship throughout the transitional stages 
when the divine spirit first began to be represented in human form . . " 

It would not be right to say of the Maori, or indeed of any 
Polynesian, that they worshipped the axes. Polynesians did not 
worsliip '* sticks and stones " ; but they addrCvSsed the god or spirit 
which for the time being, and at the call of the priests, took up their 
residence in such things as these sacred axes, and it is possible this 
was the attitude of the Minoan people of (^rete. This is born out by 
the following sentence copied from the article ** Crete Archeeology,'^ 
in the 11th Edition of the *' Encyclopedia Britannica,'* by Dr. Arthur 
Evans. He says, '* Trees and curiously shaped stones were also 
worshipped, and artificial pillars of wood or stone . . . The essential 
feature of tliis cult is the bringing down of the celestial spirit by 
proper incantations and ritual into these fetish objects . . . ." This 
exactly describes the attitude of the Polynesian towards their so-called 
** idols,'* of whatever form. 

Coming back to the Pacific, we find in Mangaia Island of the 
Cook Grou[>, stone axes with the most elaborately carved handles, of 
such a form that they never could be used for practical purposes. 
They are entirely ceremonial, and were formerly looked on as 
possessing god-like powers. Some very fine illustrations of these axes 
are to seen in a paper by Dr. Hjalniar Stolpe of Stockholm, entitled 
** Evolution in the Ornamental Ajrt of Savage Peoples," a translation 
of which, made by the wife of our life mem])er Dr. H. CoUey March, 
was published in ** The Transactiims of the Eochdale Library and 
Scientific Society " (year not given). 

Mr. Elsdon Best in Bulletin No. 4 of the Dominion Museum, in 
** The Stone Implements of the Maoris," has some notes on the sacred 
character of stone axes. The sacred character of certain stone axes 
seems to be a wide spread belief.* 

♦ A reference to ** proceBsional axes '* in New Guinea may be quoted here : 
See p. 334 of ** Work and Adventure in New Guinea," by the Rev. J. Chalmers 
and Rev. W. Wyatt GiU. 



Digitized by 



Google 



TRADITIONS OF AND NOTES ON THE 
PAUMOTU OR (TUAMOTU) ISLANDS. 



Collected by the Rev, Phe Hervi Audran^^ of Fakahina^ 
Paumotu Islands, 

TRANSIATBD FROM THE FHBNOH BY B. H. ROOKEL, M. A. 



PART V. 

(Continued from Vol, XXVlll.^page 167,) 

FORMER POLITICAL CONDITIONS AT FAKAHINA. 

IN the remote past the island was divided into three distinct districts 
peopled by three families known iinder the names of (1) Gati-Tane 
at Gake ; (2) Gati-Maliinni at Raroand atKereteki; (3) Gati-Tekopu 
(this family probably emigrated to Hao). These three tribes were 
almost always at enmity and often at war amongst themselves. If by 
any mischance a member of tlie Gati-Tane living at Te Matahoa 
ventured to enter the territory of another tiibe, he would certainly be 
seized and immediately sent to the ovens ; sucli, too, would be tl»e fate 
of any member of another tribe venturing to enter his district. A 
trespasser was a prisoner of war. On principle these unfortunates 
were never spared. The Polynesians knew no mercy so far as the fate 
of these victims was concerned. According to the testimony of early 
navigators and of the natives to-day the population of the group was 
numerous, more vigorous, more industrious, less prone to error and 
more moral than it is to-day. These facts contradict tlie law of 
continuous progress invoked by those scientists who claim for them- 
selves the honour of descent from a simian ancestor, but they are facts 
all the same. That the race which now inhabits our islands has 
degenerated more or less imder the influence of isolation and of the 
errors of unbelief, everything, including the beauty and richness of its 
language, goes to prove. But the most powerful proof lies in the 
thousand legends which we are gradually discovering, and which are 
revealing to us the ancient traditions. The investigator is surprised 
and delighted to find in the midst of a mass of incoherent stories the 
idea of the immortality of the soul, of judgment after death, of hell, 

* We deeply regret to say that the author of these papers fell a victim to the 
iiiflueiiza epidemic of late last year. We have reduced this paper considerably as 
it contained matter outside that ordinarily published by the Society. — Editor. 



Digitized by 



Google 



Traditions of the Paumoiu Islands. 233 

the idea of sacrifice and of prayer, and tlie memory of a beneficent 
superior being who lias created men and who, curiously enough con- 
sidering the prevalence of polygamy, gave the first man one wife, and 
one only. ^ 

EOYAL DESCENT IN NIUHI. 

The ancient history of the Panmotu Islands is in utter c<mfusion, 
and the most experien.ced annalist would find himself helpless and at 
a loss in dealing with the subject. The elders liave disappeared, 
taking into tiie grave witii them the traditions and the greater part of 
the ancient songs, which are the sole source for a written history in 
these distant isles where writing was unknown. To attempt, then, to 
reduce the history of those times of change to the systematic regularity 
of present times would be to attempt the impossible. It is useless to 
think of establishing a genealogy of kings of these different districts in 
the male line of descent, a line of kings who ruled over the whole 
Island. This idea of a united monarchy would clearly be far from the 
reality, while to prove hereditary occupation of tlie throne would be 
an insuperable difficulty. There seem to have been at Fakahina of old 
tliree separate tribes, each with its own king : The Gati-Tane at Te 
Matahoa or Gake, the Gati-Mahinui and the Gati-Tekopu in the west 
at.Jiaro, and in the south at Kereteki. But these two latter tribes 
intermingled and lost their identity, so that while nominally there were 
three tibes in the island, there were in reality only two. The Gati- 
Mahinui and tlie Gati-Tekopu joined forces and formed in reality one 
tribe, although the Gati-Tekopu supplied several distinguished kings, 
amongst these being Maitupoa, Tagihia, Tu-Garue and Maruake. The 
latter was chief or king when the Paiore incident took place. The 
following is a list of kings who have reigned in Fakahina : — 

I. — In the tribe of Gati-Tane. II. — In the tribe of Gati-Mahinui. 

1. Tane-tupu-hoe 1. Mahinui Te Tauira 

2. Te-tohu 2. Marere-nui 

3. Toa-rere 3. Te-fakaliira 

4. Te-mapu 4. Eogo-te-kapu 

5. liua-kai-atua 5. Tai-te-ariki 

6. Te-ata 6. Tehu 

7. Tane 7. Mahinui 

Etc. Etc. 

If an average of twenty-two years ])e allowed for each reign, 
Fakahina has had kings only for a period of 154 years. 



1 . I intend flhortly to undertake a study of the numerous points of contact 
between Polynesian traditions and our knowledge of revealed religion. 



Digitized by 



Google 



234 JOURNAL OF THE POLYMESIAM SOCJETY. 

BBLIGION IN FAKAHINA: MARAISS, TAHUAS, OFFICIATING PRIESTS 
AND THEIR JURISDICTION. 

In pagan times Fakaliina coiitained many mara^^ the six chief 
being: — I Aehaii ; 2 Oroniea; 3 Ragi-te-tau-noa ; 4 Pekai ; 5 Vai- 
toinoana ; 6 Tugata. 

In the Tuamotu Group the chief officiating priest, who conducted, 
so to speak, divine worship, and represented the arclipriest in our 
cathedrals, was known as the kaunuku. He was a great personage 
and very holy. Further, he enjoyed the highest privileges. He was 
exempt from ordinary work and from that forced labour at times so 
troublesome, such as cooking and the preparation of the turtle, for 
which the common people were liable. The smoke from the ovens 
was not to come near him or to touch him. Throughout the whole 
island there was but one authority (that of the king) superior to his, 
while at times his influence was as powerful as even that of the king. 
He alone was responsible for the ordering and carrying cmt of eveiy- 
thing that ctmcemed the celebration of the annual festivals and the* 
performance of the religious ceremonies on the marae. All these were 
under his sole jurisdiction. It was i\\Q kaunuku whose duty it was to 
regulate them as he thought fit, provided that he preserved the ancient 
usages. To the high-priest belonged by right the privilege of taking 
from the fare-tini-atua, corresponding to our tabernacle, the sacred 
stone, and laying it on the turtle for some minutes before cutting its 
throat. A few paces distant from the kaunuku stood two other 
officiating priests called huhuki, who were his assistants and, so to 
speak, the deacon and sub-deacon, and consequently of subordinate 
rank. These latter were not the ordinary priests, but were of royal 
blood. When the turtle was divided up, the head, pepenUj the fat 
part tliat surrounds the neck, genegene, and the heart, mafatu, were 
always set apart and reserved for the kaunuku. The children, young 
people, old men and the women were rigorously excluded from the 
marae. It was only those of maturer age, %^y. thirty years or so, wlio 
could cross the sacred enclosure, take part in the ceremonies of the 
cult, and have the right and permission to eat of the turtle. 

FAMOUS NIUHIAN VOYAGES. 

The islanders still retain tlie memory of several famous voyages. 

1. Mahere : — The undoubted cause of many Polynesian voyages 
and migrations was the quarrels stirred up on account of women. 
This was the case with Marere as with many others. He had relations 
here with Te-kopu-hei-ariki, a woman of royal blood, and from these 
relations were born two children, Te-fakahira and Tu-te-ragi-nui. 
On the birth of her son Te-kopu-hei-ariki raised him aloftj according 
to an ancient custom among the nobility, at the same time singing a 
new song, of which the following words are still preserved ; **Nafea 



Digitized by 



Google 



Traditions of the Paumotu Islands, 235 

-vau o Te-kopu-liei-ariki toku ariki." Shocked at liearitig these words 
from his wife, and in anger at being deceived in her character, he 
-went on board his canoe and fled to Takunie. There he foi-nied a 
•connection with anotlier woman named Te-pogi. By her he liad also 
tliree children, Puraga, Te-taukupu and Te-fau. But a matrimonial 
•experience similar to his former one caused him to leave Takume. 
He made for Fagatau, where he made a third alliance, this time with 
a woman named Mahuru, by whom he had a child named Yaroa. 
The name of Marere's canoe is unknown to-day. 

2. Tehu : — ^This man has remained famous in Fakahina not only 
l)eeause of his numerous long voyages of exploration, but chiefly 
because he presented his country with fruit-trees and food-plants. 
As a matter of fact it was he who, on his return fi*om one of his 
voyages on board of his ^'Katau," introduced the coconut, the taro^ 
the ape^ etc., into the island. Thus he has become the great bene- 
factor of his countrymen, in whose memories his name is deeply 
engitived. 

3. Te-fakahira : — Doubtless inheriting the qualities of his father 
Marere, and especially his love for long and perilous voyages, Te- 
fakahira too traversed all the seas of the archipelago, and visited 
Takoto, Re-ao and Hao without reckoning the thousand other islands 
unknown to us. At Hao he became the father of a son named Te- 
mauri, whom, on his return, he brought back home with him. 

4. Te-mauri : — After liaving grown up at Fakahina, and, like his 
father, smitten with a love for travel and adventure, he set out. He 
went first to Marekau, where, according to the story, he had three 
wives, Takua,"Vero-niatau-toru and Te-rapure-ariki. He then went 
on to Hao, where he became acquainted witli Gahina. 

5. Faruia : — According to tradition this man went as far as 
Yairatea. He was a colossus and an athlete. He lifted his canoe on 
to the land, and, witli the assistance of his crew, he killed all the 
inliabitants of the island. It was, no doubt, to avenge this massacre 
that the inhabitants of all the adjoining islands formed an alliance 
and, on board of seven canoes, came and made war on him. But he 
defeated them and killed most of them. He had with him his old 
father, whom he carefully hid in the tap-roots of a paudanus, strictly 
charging him not to come out of this hiding-place. But the unfortu- 
nate old man hearing that his son was scattering the ranks of his 
enemies, and that he had killed most of them, moved by compassion 
or some other feeling, came out of his hole. Misfortune soon overtook 
him, for some of the enemy in their flight chanced to meet him. He 
was instantly seized and put to death without mercy. As soon as 
Faruia heard of this he hastened to the bleeding and still warm body 
of his father. Stricken with grief he threw himself on the body and 
gave vent to his sorrow. His enemies taking advantage of the 



Digitized by 



Google 



236 JOURNAL OF THE POLYNESIAN SOCIETY. 

attitude into which ]iis infatuation had thrown him, killed him in his 
turn. 

6, Makuake : — Tliis navigator heh>ngs to more modern times* 
He went to Takoto and seized the wife of Porutu. Because of lii» 
desire to take her off with him to Fakahina, or elsewhere, Porutu in 
anger struck him a severe blow on tlie head. The would-be ravisher 
returned to Fakahina in his wounded condition. There by means of 
cures and frequent washings he succeeded in healing his broken head. 
Three wells or water-holes, Te-vai-marigi, Te-vai-totoniea and 
Tamunu, where he performed his frequent abUitions, are still pointed 
out. Naturally he now had but one idea, to take venj^eance on his 
enemy for tlie wound he had inflicted, by killing him. To accomplish 
his aim he went to Takume and to Rairoa to obtain hel[>, and tlien stood 
in for Takoto. He killed many of the islanders and among them 
Porutu. He seized the woman once and for all, and returned to 
Fakahina with her. 

The following are the most famous canoes of the Niuhians : — 
1 Pua-te-nukuroa ; 2 Te-vai-tau ; 3 Houpo ; 4 Maramu; 5 Tupou ; 
6 Takau, whose privateer-captain was the celebrated Tehu ; 7 Tekoro, 
which had as its coxswain Te-ata; 8 Torona, Heko, Te-rate-maro^ 
Te-piri. 

THE FIKST VISITORS TO FAKAHINA. 

1. The first pahi (or ocean going canoe) seen at Niuhi, and one of 
which there still exists a dim recollection, came from the Marquesas. 
This was a Nuku-Hiva pahi, according to the natives. The tradition 
relates that only one person leaped into the water and came ashore 
by swimming. This was a woman named Mahuru. Some of the 
islanders wished to kill her, while others wished to spare her. The 
latter proposition prevailed, and thus she was saved. Some days later 
her relatives came to seek her, and she was able to be placed once 
more on board her own canoe. 

2. Such, unfortunately, was not the fate of the Paumotu pahi 
commanded by Manava-rere. This canoe came from the west. The 
whole numerous crew of fifty were, it seems, massacred and decapi- 
tated. The bodies were buried in the marae of Katipa, situated 
beside the open sea, while the heads were hidden in the marae of 
Oromea, close to the lagoon. Ua taparuhia ratou e to uta. "They 
were allured by the dwellers of the island." 

Thi« slaughter was carried out by the orders of Te-ragi-heikapu^ 
the ruling chief, who directed the operation in person. Manava-rere 
besides being an experienced mariner was a man of great physical 
strength. In this respect he was a worthy match for Te-ragi-heikapiu 
It seems that when he was about to be captured beside the open sea, 
he gave a tremendous leap and found himself on the edge of the 



Digitized by 



Google 



Traditions of the Pauihotu Islands. 237 

lagoon, and tlien from there on to tlie offing. However, lie was 
iinally seized hy his maro and l)Ound. He was securely tied to a large 
8tone lying in a small lake, nakana, where his enemies tried in vain to 
kill him. Every morning they went to see if he were not dead. 
Now, on each occasion they found him still hreathing. One fine day 
lie said to them : *^Aita van i higa ia koutou ; te tagata ra i higa vau i 
ana o te tagata i tapu i toku gogo pito naku i tapuV " You cannot kill 
me, except by cutting my umbilical-cord ; " an expression, which in 
their language, is equivalent to saying that he himself belonged to 
Fakahina. Therefore he was spared, and, in due course, became the 
father of a numerous line, which reckons no fewer than ten 
genei*ations to the present day. 

3. The Takahi, a canoe from Fagatau, had a similar experience. 
As the inhabitants of Fakahina and of Fagatau are related, the latter 
were not molested, but, on the contrary, met with a fiiendly reception. 

4. First visit of Paiore to convert the Tuamotu peoples to civilisation. 
On his first visit Paiore was well received at Fakahina. He gathered 
the whole population together, and remained with them two or three 
days. He was entertained in native fashion, and went away delighted 
with his visit. 

5. Second visit of Paiore in 1860. Paiore's second visit ended in a 
tragedy. His crew was composed of ten or eleven men collected from 
various places, for he had with him a representative from nearly all 
the other islands. On this occasion, contrary to his behaviour on his 
first expedition, Paiore did not go on shore, but allowed or even ordered 
seven of his men to land. This was indiscreet, and at the same time 
a grave blunder. He must have seen this afterwards, but then 
it was too late. Six of the seven men sent on shore were killed. 
Their names and birth-places are as follows : — 1 Tapahiha of 
Fakarava ; 2 Mahiri of Makenio ; 3 Taumata of Taenga ; 4 Tu- 
ata of Nihiru; 5 Te-hei of Takoto ; 6 Tahoro of Reao. The 
seventh was Turia of Makemo, who owed his escape to his fleet- 
footedness. Throwing off to his pursuing foe also his vest and hat, 
he managed in his mad flight to Gake, to outdistance his pursuer, 
throw himself into the sea, and reach Paiore^s cutter by swimming. 
He, as an eye-witness and a participant in the bloody affair, related 
the whole story with all its horrible details to Paiore. Under his 
very eyes Paiore's sailors were pursued, captured and killed, one after 
the other, on the reef. 



Digitized by 



Google 



238 JOURNAL OF THE POLYNESIAN SOCIETY. 

ANCIENT NIUHIAN SONGS. 



FAATARA NO TB ARIKI RA MAHINUI TE-TAUIRA E TIA I FAKAHINA. 

Fauau a raro, ko Mahituii Te-Tauira : ko Niuhi kiukiu, nauiiau^ 
ragitaka mahiti ake i raro te aka o te iienua, ka tagi te pahu ko te 
rutu ko Mapuna iavaku ki mua i o Kogo. Hopukia te moana, kauria 
i te uioana, te moana uriuri, te moaua kerekere, ka biga ki uta i te 
henua, lieuea, te papa ka gatata, Porutu tautua, Porutu tauaro. 

Kopekahaga ia o Te-Kura ma te molio ki mua i o Rogo, topa te 
pin o Mauogi, tiraga ruperupe, ki reira, vaitorotoro knriri tagihoro e 
tagi te avaroa, e tagi te talia o Vavau. 

Kapu ko teie Ariki. Taku talma Kiipakupa, ko te taliua ko te 
vai marigi Maregai ko keha. Tika i a mari Tagaroa ki te tua o te 
ragi, kapu ko teia Ariki. 

Ara mapuhia, Taketake mai Hiva, kapu ko teia Ariki. Mau atu 
to maro ki mua i a Hau, koutu e tere, kapu to teia Ariki ko Mahinui 
Te-Tauira, turuturu ki mua i o Rogo ka haruru tana kopu, mi tiki a 
tana ta liaohoa, ka ririu, tua ka ririu aro. Maeva, ko te pu ko te paliu^ 
ko te raukava ka pupulia nukunuku ma te fakiteragi matereua. 
Maeva te Ariki. Maeva Te-uho te Ariki ko Mahinui Te-Tauira. 

FAATARA NO MARUAKE. 

Karilii e karilii nui a kae koi te matau o Rogo fatia fatia e, tara e 
tara, taliuri mai kona tena te potiki, toa e toa e, toa e. Te lietu ma 
te marama tena potiki, toa e, toa e kapu korero kapu vanaga tena 
potiki, toa e toa e, toa e. Kapu iti kapu ai, tena te potiki toa e ko 
Havaiki, ko Havaiki. 

Havaiki tinilii koi ruga, Havaiki tinihi koi raro, piri mataitai ma 
te nariki, te liuru o Te-kura ma te nariki, te mata o Te-kura ma te 
nariki, te hope o Te-kura ma te nariki, piri mataitai ma te naiiki, 
maro kapu koi ruga, maro kapu koi raro, maro e kapu e Ariki. 

FAATARA NO HOG A TATAOA. 

Fanaua i raro ko Hoga Tataoa, ko Niuhi paka koru, takurua ki 
te hekeheke ko purepure, i hiti kapn ko te ie Ariki. Ko te iho, tena 
kabapu, uiai kona, ka turaha mai kona, ka purero mai kona, ka haere 
mai kona, ka haere mai kona, ko purepure i hiti, ka fanau ai te Ariki 
Tataoa. 



Digitized by 



Google 



Traditions of the Paumoiu Islands. 



239 



Tlie following list of the all too scanty flora of Fakahina may be 
of interest both to the pliilologist and to the botanist : — 



TREES AND PLANTS. 



Ancient Paamotu 


Modem 




Language. 


Names 


Scientific Names. 


Niu 


Hakari 


Palma nucifera 


Maliame 


Gatae 


Pisonnia ombellifera 


Piupiu 


Geogeo 


Touniefortia argentes L. 


Kokiirii 


Uu 


Suriana Maritima L. 


Tou 


Tou 




Gagie 


Mikimiki 


Pemphis acidula Forster 


\ Viri . 


Tima (fara) 


Nono 


Hora 




Putamagomago 


Putarau 


Sesbania grandiflora, Pers. 


Goliegohe 


Kikipa 




Koporoporo 






Gapata 




Scevola konigi 


Kama 






Kahia 




Guettaada speciaso 


HERBS. 






Mauku 




Lepturus respens R.B. (Gramine 


Nanamu 






Mahihi 






Buna 






Parahiralii 




Heliotropium anomaluni H. et A 


Toroariki 


Gaio 




Nau 


Horahora 


L Lepidium piscidum 


Vaianu 


Ogaoga 




Kainoka 






Maukutoga 




THE FAUNA. 


THE TURTLES : 






Common Turtles. 


Rare Species. 


Kurahiva 




Topitopi 


Marega 




Tumi mi Mokamoka 


Kea 




Kogaga Paku 


Gaki 




Purekau Totoro tika 


Igoa-kore 




Kagihau 


Maunii 




Tokau 


Tauhetika 






Heka 







Digitized by 



Google 



A MAORI STONE AXE. 



By H. D. Skinner. 



THE stone cutting tool figured in the plate, now in the collection 
of the British Museum, still has attached the printed label of 
the New Zealand Geological Survey, on which the locality is stated as 
** near Wellington." It was presented to tlie British Museum, 
together with a number of other stone implements bearing the same 
label, by the late Sir James Hector, F.R.S., then Director of tlie 
Survey. 

Its special interest lies in the fact that it can have been used in no 
other way than as an axe, whereas there is nothing in the shape of 
the only other type of Maori stone axe thus far described* to prevent 
its former owner from hafting and using it as an adze, if no true adze 
were available. The distribudon of this new type of axe is interesting. 
A good unfinished example from the Lukin*s collection is in the 
Nelson Museum, and was found, I believe, at D'Urville Island. The 
material and technique of the British Museum specimen render it 
likely that it was made in the same district. Mr. Elsdon Best tells 
me that he has seen a greenstone example found in the neighbourhood 
of Wellington. An example in the Canterbury Museum is, as Mr. 
Speight informs me, almost certainly from the Canterbury district, 
while an unfinished tool of the same type, found near Cape Saunders, 
Otag'o, is in the collection of Miss B. Howes. Thus three known 
examples come from the South Island, and the remaining two from 
the extreme south of the North Island. The type does not seem to 
have been recorded in any other part of the Pacific. 

Maori stone axes previously described appear to belong to an 
entirely different t3rpe, being designed, so far as can be judged, for 
insertion either in the slot in the cleft foot of a wooden haft (Fig. 3), 
or against the side of the foot, which was specially flattened for the 
purpose, t 

Figure 2 indicates the relation of stone implements to wooden haft 
in the case of such tools as Figure 1. Figure 3 indicates the same 

♦ Elsdon Best: ^^rlomimoii Bulletin," No. 4, pp., 136-162. 
t Best, loc. cit., p. 137, and Plate XiiH. 



Digitized by 



Google 



gitized by 



Google 




Digitized by 



Google 



A Maori Stone Axe. 241 

relation when the stone axe, which in this case was of an entirely 
different type from Figure 1, was inserted in a slot. The method of 
binding has not been indicated in either case, as I have not seen old 
examples in wliich the, fibre has been preserved. The haft of Figure 
2 is drawn from an ancient example dug up at Otokia, Otago, and 
now in the Otago University Museum. That of Figure 3 is based 
on a North Island adze handle in the same museum. The setting of 
the 8t<me axe in the slot is copied from a Taranaki example now in a 
private collection, which, though made by a Maori in quite recent 
years, probably follows the old method. The stone axe in the 
Taranaki example was polished all over, and very closely resembled 
the type of adze commonest in that district and in the rest of the 
North Island. Such axes seem always to have been classified as 
adzes until Mr. Best's researches proved their real nature. It is quite 
possible that the Maoris sometimes hafted and used such axe-blades 
as adzes, for a large class of stone implements from New Guinea 
resembling this one in shape were used in either way, as occasion 
required. 

DESCRIFriON OF FIGUKES IN PLATE. 

Figure 1 (A, B and C). — Stone axe. Loc. near Wellington, N.Z. 

Scale i. British Museum collection. 
Figure 2. — Position of axe on haft : southern type. 
Figure 3. — Position of axe in slot in haft: northern type. 



Digitized by 



Google 



NOTES AND QUERIES. 



[201] The Three Fingers in Maori Carvings. 

We notice in Mr. H. B. Cotteriirs ** Ancient Greece,*' 1913, a picture of 
what one might call a ** plaque/' found in the excavations at Knossus, Island of 
Crete, depicting, according to the title of the picture, *' Genii (Priests?) watering 
a sacred tree." What will intereHt Maori scholars are the two Bgures of the 
? priests, who have each three fingers to their hands and three toes on each foot. 
We need hardly call attention to the fact that on Ancient Maori Carvings the 
human figures, in nearly all cases, have only three fingers. The same three fingers 
have been traced on carvings in India. What is the origpinal meaning of this 
peculiarity ? 

The Minoan civilization of Crete also, apparently, recog^sed the supernatural 
powers of certain axes — double-headed in those caset*, and called Labrys — as did 
the Maoris. Both the Minoan and Polynesian peoples belong to the Caucasian 
race. 

Editob. 

[292] Polynesian Voyages. 

In that excellent publication " The Museum Journal " for March- June, 1919, 
issued by the Museum of Philadelphia, we notice a paper, with an illustration, on 
a South Sea Native Chart, derived from the Marshall Group of islandH. The chart 
does not differ much from one described and pictured in *' Hawaiki," p. 187 
(third edition). But the author of this paper makeH one statement which it is 
necessary to correct. He says the Marshall Islanders made the most extensive 
voyages of any of the South Sea peoples— and some of their voyages extended to 
600 miles. A reference to the firMt paper in this number of our '* Journal " will 
show that the distance of 600 miles has been exceeded by the Polynesians over and 
over again. One voyage is thus mentioned, that of Tangiia from the Fiji Group 
to Easter Island, is 4,200 miles in a straight line. Many voyages have been made 
by the Polynesians from Tahiti to New Zealand and back, in the tenth to the 
fourteenth century, and the distance that separates these two places is about 
2,400 miles. The Marshall Islanders are Micronesians not pure Polynesians. 

Editob. 



Digitized by 



Google 



PROCEEDINGS. 

POLYNESIAN SOCIETY. 



The Council met at the Library, Hempton Rooms, on the 3rd December, 1919, 
when there were present : The President, and Messrs. Bullard, Newman and W. 
W. Smith. 

After the minutes of the preceding meeting a letter from Mr. J. B. Roy was 
read resigning his seat on the Council. It was resolved that the Council receives 
with regret Mr. Roy's resigfnation, and tenders its warm thanks to him for his 
seryices for the last six years. 

It was i-esolyed that Mr. W. H. Skinner be appointed as a member of the 
Council. 

Letters were read from Mr. Redcliffe Brown, of Tonga, resigning his 
membership on account of ill-health. Also one from Dr. Gregory, of Hayard, 
announcing the news of an expedition to study Polynesian and other matters in 
the Pacific under his direction, and at the cost of the Pauahi Bishop Museum of 
Honolulu. From Mr. Thos. G. Thrum, of Honolulu, with thanks for his 
appointment as a Corresponding Member. 

The following new members were elected : — 
Mrs. Ann Elmsley, Hill Side, Waverley. 
J. Corlett, P.O. Box 38, Taumaru-nui. 
P. Alfred Grace, Tokaanu, Taupo. 
William Cooper, Gisbome. 

The TumbuU Library, Bo wen Street, Wellington. 
Ian Roy, c/o J. B. Roy, New Plymouth. 
Papers received : — 

Nine Folk-lore. By G. N. Moriis. 
Notes on the Paiunotu Heavens. By J. L. Yoimg. 
It was reports that the ** Bulletin of the Societe d' Etudes Oceanienne " of 
Tahiti, announces the death of two of our Corresponding members, the Rev. Pere 
Audran, and M. Tati Salmon, through influenza. 



We notice by tlie '* New Zealand Gazette,'* of 20th November, that the 
following members of this Society have been honoured by the inclusion of their 
naines in the list of the first twenty Fellows elected by the New Zealand Institute. 

Mr. Elsdon Best, Hector Medallist. 

Dr. Patrick Marshall, M.A., Dr.Sc, F.G.S. 

Mr. Stephenson Percy Smith, F.R.G.S., President Polynesian Society. 

Dr. James Allan Thompson, M.A., Dr. Sc, A.D.S.M., F.G.S. 
The Society offers these gentlemen its congratulations on the honour conferred. 



Digitized by 



Google 



Digitized by 



Google 



INDEX TO VOL. XXVIII. 



Compiled by W. H. Skinner. 



Annual Meeting of Society, i 
Annual Beport of Council, i 
Atia-te-vannga. The most ancient land 

known to the Rarotongans, 64 
AuDRAN, Rkv. PURE Hbrvr. Traditlous of 

.and Notes on the Paumotu Islands, 31, 

161, 232 
Au-roroa. Original home of Polynesian race, 

19 
Avaiki (Hawaiki). Tangiia voyages to, and 

interviews the gods, 189 
Axes. Maori belief in the supernatural 

powers of certain axes, 229 
Ayran and Polynesian points of contact. By 

8. Percy Smith, 18 

Balance Sheet of Society for year ending 31st 

December, 1918, iv. 
Bkattik, H. Traditions and Legends of 

Southland, N.Z.. 42, 152, 212 
Best, Elsdon. The Land of Tara (Wellington, 

N.Z.), 1, 79, 123 
Bewitched trees, or sacred trees, 218 
Burial rites of San Ciistoval (Solomon Islands), 

39,103 

Coconuts of Paumotu Group, Eastern Poly- 
nesia. Varieties of, 167 

Combat between Moeava and Patii-a at 
Makemo, Paumotu Island, 32 

Dentrecasteaux (French Navigator). Visit of 
to the North Cape, New Zealand, March, 
1793, 117 

Double Canoes. South Island, N.Z., 157 

Easter Island. Account by James Gilbert, 
Master of the " Resolution," 1774, 178 

Easter Island. Visited by Tangiia-nui from 
Fiji Group, 13th Century, 192 

Exchanges. List of and publications received 
for 1918, X., xii. 

Fairies. Myths of, 28, 64, 215 

Fakahina Island, Paumotu Group. Discovery, 

description and traditions of, also religion, 

poetry, famous voyages, etc., 161, 232 
Fatherland of the Polynesians. A Rarotongan 

description of, 61 
Fights. A list of engaged in by Ngai-tahu and 

Ngati-Mamoe tribes. New Zealand, 45 
Fox, Rev C. E. The San Cristoval (Solomon 

Islands) Heo (Burial Mound), 39 

Gilbert's Account of Easter Island. Capt. 

Cook's 2nd voyage, 178 
Graham, Geo. Rangi-Hua-Moa, A legend of 

the Moa, 107 
Graham, Geo. The account ot Kupe and 

Tainui, 111 

Hine-nui-te-po. The goddess of Hades, 160 

Jonah and the Whale. The Pacific story of 
(Note 289), 121 

Karakia or Invocation used in planting the 

kumaray 27 
Karika, a great Polynesian Navigator of the 

13th century, 195 



" Koro-tuatini." Name of the great temple 

in the Polynesian Fatherland, 64 
Kumara tuber. The. Ancient myth eis to its 

oiigin, 26 
Kumu-honua. The Hawaiian name of the first 

man, 56 
Kupe and Tainui. The account of arrival and 

discoveries in New Zealand, 121 

Leverd, M. Armand, The late. Correspond- 
ing member at Tahiti. Photo of, 122 
Lizards, or Ngarara. Stories of, 217 

Maeroero. Mythical wild men of the woods. 

A hairy monster, 212 
Maketu Island of the Paumotu Group. Visited 

by Tangiia the voyager in the 13th century 

196 
Malayo-Polynesian (Note 288), 121 
Man. The origin of. Rarotongan Myth, 56 
Maori belief in the supernatural power of 

cei-tain axes, 229 
Maori stone axe. By H. D. Skinner, 240 
Matariki, or the Pleiades star group. Used 

by Polynesians to mark commencement of 

the New Year, 27 
Members of Society, 1918. List of, v. 
Moa (The). Methods of hunting by the old- 
time Maori, and description of bird, 108 
Moeava. An early Polynesian navigator. 

Traditions of, 32 
Morris, G, N. Nine Islanders Folk-lore, 226 

Niue Islanders. Folk-lore. By G. Morris, 226 
Notes and Queries, 54, 121, 180, 242 

Pai)a-tua-nuku. The Earth-mother. Separa- 
ted from the Sky-father, 229 

Parata. A monster which causes the tides, 30 

Pare (Ornament over doorway). An ancient 
carved. Auckland, N.Z., 160 

Paumotu (or Taumotu) Islands (Eastern 
Pacifiq). Traditions of by Rev. Pere Herv6 
Audran, 31, 161, 232 

Paumotu. The Conception of the Heavens 
and of Creation. By J. L. Young, 209 

Polyriesians. Their first thi-ee migrations into 
the Pacific, 55 

Polynesian .Linguistics. (Santa Cruz Archi- 
pelago). By Sidney H. Ray, M.A., 168 

Polynesian Voyages. (Note 292), 242 

Proceedings of Polynesian Society, 54, 122, 181, 
243 

Rangi-Hua-Moa. By Geo. Graham. A legend 
of tho Moa, 107 

Rangi-nui. The Sky-father, separated from 
the Earth-mother, 229 

Rapa-nui (Easter Island) and Rapa-iti, refer- 
ence to, 178, 192 

Rarotonga. History and traditions of. By Te 
Ariki-tara-ara, 55, 134, 183 

Ray, Sidney H., m.a. Polynesian Linguistics 
(Santa Cruz Archipelago). 168 

Rongo-ma-Uenga, the god. Stolen by Tangiia 
at Tahiti, 189 



San Cristoval (Solomon Islands) Heo, The. 
Rev. C. E. Fox (burial rites), 39, 103 



By 



Digitized by 



Google 



-246 JOURNAL OF THE POLYNESIAN SOCIETY. 



•€lanta Cruz Archipelago, Western Pacific. Dis- 

covery) languagre, etc. By Sidney H. Bay, 

M.A., 168 
^AVAOB, Mb. S. Histoiy and traditions of 

Barotonga. Translated by, 67 
43oalp-taldng, Very like. By 8. Percy Smith, 

106 
■Science Congress. Ohristchurch, N.Z. Papers 

on Polynesian Ethnology, 52 
Skinner, H. D. Gilbert's account of Easter 

Island, 1774, 178 
Skinner, H. D. A l^Iaori stone axe. With 

plate, 240 
Smith, S. Percy. The Fatherland of the 

Polynesians, 18 
Smith, S. Percy. Very like Scalp-taking, 106 
Smith, S. Percy. History and Traiditions of 

Rarotonga. Translated bv, 134, 183 
Swimming. Maori methods of, 221 

Taepo and Taniwha. A goblin and a water 

monster. Stories of, 214 
**Tainui*' and Kupe. The account of. By 

Geo. Graham, 111 / 

Taki-pu (afterwards re-named Taki-tumu) the 

ocean going canoe in which Tangiia-nui 

explored the Pacific— 13th century, 188, 197 
Tamarua family of Rarotonga. History of, 64 
Taniarua-pai navigates "Taki-pu," the canoe 

of Tangiia-nui, 189 
Tangiia-nui of Rarotonga. A great navigator. 

Story of his life and voyages throughout 

the Pacific— 13th century, 67, 134, 183 
Taputapu-atea, son of Whiro of Rarotonga 

(13th century) taken from Easter Island, 

192 
Tarionge of Huahme, Tahiti. An ancestor of 

the Taranaki tribe, New Zealand, 193 
Tattooing. Notes on, 222 
Tawhiti-roa of Polynesian, legend. Surmised 

to be Sumatara, 62 
Te-Amio-enua. A celebrated weapon of Tahiti, 

and the cause of wars— 13th century, 187, 

190 
Te Awhio-i-angi. A famed Polynesian axe of 

supernatural powers, 229 
Te Ariki-tara-are. The last of the high-priests 

of Rarotonga. His history and traditions 

of the Rarotonga, 55, 134, 183 
Te Aru-tanga-nuku of Rarotonga and his 

canoe and voyage^ {circa a.d. 1000), 139 
Te Irapanga. Leader of the migration from 

Tawhiti-nui (Borneo?) to the Hawaiian 

Islands, 135 
Te Nga-taito-ariki. A great ancestor of the 

Rarotongans, 134 



" Te Paerangi." Wars of the gods. Ancient 

myth, 32 
The Fatherland of the Polynesians. By S. 

Percy Smith, 18 
The Great Mum (bA Opunaki, Taranaki). By 

a Taranaki Veteran, 97 
"The Land of Tara" (Wellington, N.Z.) By 

Elsdon Best, 1, 79, 123 
The three fingers in Maori Carving. Note 291, 

242 
The Turehu, or Patu-pai-a-reke. A fair or 

white people. Faines, 28, 64 
Traditions and Legends of Southland, N.Z. 

By H. Beattie, 42, 162, 212 
Traditions and Notes on the Paumotu Islands, 

By the Rev. Pere Herve Audran, 31, 161, 

232 
Tumu-te-varovaro, ancient or spirit name of 

Rarotonga, 190, 194 
Tu-tarangi of Rarotonga and his wars {circa 

A.D. 425), 136 
Tuoro (or Puke-tuoro). A mythical water 

monster, 218 ' 
Turihuka, wife of Tamatea (a.d. 1360). Ex- 
plores Southern New Zealand, 223 
Tu-tapu, A celebrated ariH of the Marquesas 

—13th century, 185, 195 

Uea or Wallis Island, N.E. of Fiji Group. 
Visited by Tangiia the voyager, 13th cen- 
tury, 191 

Ui-te-rangiora. The great Polynesian navi- 
gator and discoverer. Circa a.d. 660, 63, 
67, 137 

Upolu, Samoa. Visited by Tangiia the voyager 
in the 13th century, 191 

Vega (the star) or Whanui. The very ancient 
"Year Star** (New Year) of the Poly- 
nesians, 27 

Veka (Weka). The, and the Kale. A Niue 
folk-lore story of these two birds, 226 

Voyages of old-time Polynesian navigators, 

Wanaka District. The. (Otago, N.Z.). Note 

290, 180 
Whiro. An ancestor of the Rarotongans, 

Tahitians and Maoris, is met in Fiji Is^ . 

lands, by Tangiia— 13th century, 192 
Whiro-nui. A celebrated axe of Polynesian 

fame jKJSsessing supernatural jwwers, 229 
Wizards [tohungas). Powers of, 216 

Young, J. L. The Paumotu (Eastern Pacific) 
conception of the Heavens and of Creation. 
Illustrated, 209 



Digitized by 



Google 



Digitized by 



Google 



i 

Digitized by VjOOQIC 



RKOisTEMfsn FOR mAmmssioif as a maqazikm. 



Vol. XXVI I L [PUBLISHED QUARTERLY,] 



No. t 



THE JOURNAL 



OF THE 



POLYNESIAN SOCIETY 

CONTArNING THE TRANSACTIONS AND 
PROCEEDINGS OF THE SOCIETY. 



Published under the Authority of the Council, and Edited 

by the President. 



No, 109. MARCH, 1919, 



CONTENTS. 

[AnthorH nre alone responsible for their respective slnienimits,) 

ANNUAL MEETING OF THE SOCIETY . . 

ANNUAL REPORT OF THE COUNCIL , . 

BALANCE SHEET 

MEMBERS OF THE SOCIETY 

LIST OF EXCHANGES 

BOOKS, ETC., RECEIVED IN EXCHANGE 

THE LAND OF TARA. By EL8D03f Best 

THE FATHERLAND OF THE PULYNESIANS. By S. Peecy Smith 

TRADITIONS OF AND NOTES ON THE PaUMOTU (OR TUAHOTU) 

ISLANDS. By Rev, Peke Hervb AtnoaAN 
THE SAN ORISTOVAL HEO. By Hev C, E, Fox . . 
TRADITIONS AND LEGENDS OF SOUTHLAND. By H. Bbattie 
THE SCIENCE CONGRESS, CHRISTCHURCH, N.Z. 
NOTES AND QUERIES 
PROCEEDINGS 



PAGK 



xll 

1 

18 

31 
39 
42 
52 
54 
64 



new Plymouth, R.Z* 

PRINTED FOR THE SOCIETY BY THOMAS AVERY, 
AGBNT FOR AUSTRALIA : ANGUS and ROBERTSON, 89 CASTLEREAGU STREET, SYDNEY. 



1919, 



Digitized by LjOOQ IC 



POLYNESUN SOCIETY. 



OFFIOEKB FOR 1919. 
fnttons: 

'Thk Right Hon. Lord Pldnket, Lato Govt?njor of New Zenknd. 

Tub RioKT Hon. Losn Islington, P.Cm K.C.M»Q.* 

Latti Governor of New Zeabmd. 

This Rioht Hon. Thb Ejlrl of Livkbikjol, M.V.O., O.CMSi., 

QoTemor-Qeneml of New ZeitlHtitl. 

f mibcut : 

S. Pkrcty Smith, T.R G.S. 
fjit/ro E4Um' ofjmntal.) 



€oujkU : 



W* L. Newman, 
. J» B, Rot. 



P. J. H. White. 
W, W. Smith. 
Q, H, Btjllahd. 



W, W, Smith and W. L. Nbwman. 



T*HE Society ia formed to promote the Htndy of tbo Anthropology, Ethnology, 
* Philology, History, and Autiqiiitif?}* i)f thfi Ptjlynesiun racee by the puhli* 
cation of an offfeial journal to be calliid ^^The JotniKAi* OF run Poi^yn£SU17 
Society/' and by the colkction of books, manimtTipt*!, photogfraphs, reliog, &i>d 
other illustrations of the hist<ny of the PnlyneMiftTi race. 

Th© term " Polynetaa *"■ i^ inleiuleO to imiliide Ati»lrtila&ia, New Zealand, 
Melanesia, Hicroneffla, and Malaysia, as well as Polynettiii proper. 

CandidaieA for admisftioii to the Society shall be admitted on the joint 
recom mentation of a member of the Soeie?ty and ii meniher of the Council, and cm 
the approval of the Council. Formn of Application oan be obtained from tiie 
Hon. Secretary. 

Every persoo elected to memher**hip **httll receive immediate notice of the 
»ame from the SeeretJirie*!, together with a copy of the Rule^i, and on payment 
of his aubHCiiption of one pound shall be entitled to all the benefitB of metnhersiilp. 
Subscriptiontt are payable in advance, i>n the J*it of January of each year, or on 
election. 

Papers will he i-ef^eived on any of the abriive Buhje^it** if sent through it member. 
Authors are requested to writ* only on one nido of the paper, to tJNe quarfo paper 
and to leave one inch margin on the left-hand side, to allow of binding. Pr 
names sboiild be written in ROMAN TYPE. 

Tlie price of hack numbers of the Journal ♦ to member*i, in la. 3d. 

Members and exchaii^es are requested to note that the Socic, 
Office is at New Plymouth, to which all communications^ boot., 
exchanges, etc.) should be sent, addressed to Hon. Secretaries. 



Digitized bv 



Google 



Publications or Polpnesian Socletp. 



Thk Journal is published in the first week of January, 
April, July, and October, and contains abont 60 pages 
each issue, with occasional ilhistrations. 

A few copies of the seciond edition of Volumes IV,, 
and V* (189*2-6), price ten Mhillings a volume, are now 
available, the others arc out of print 

Volumes VI L to XXVIL are all available at five 
shillings a volume* Sold only to members of the Society- 
The volumes are unbouudj and in numbers as published 
quarterly. Postage additional, fourpence a volume. 



Memoirs and Reprints of the Polynesian 

Society. 



Vol. L — ^'HlSTORT AND TRADITIONS OF THE 

Taranaki Coast," 8vo*, cloth, 
561 pages, witli a number of 
illustratious and maps. Price, 

Vol. IL — *' The Moriori People of the 
Chatham Islands : Their His- 
tory and Traditions," Svo,, 
paper, abont 220 pages, illus- 
trated. Price, 

VoL IIL — '^The Li^he of the Whare* 
wananga.'' — Part I. : ' Thiugi^t 



10s. 6d. 



5s. 



pages. 



Uele8tiaL' About 200 
Cloth bound. 

Vol. IV. — Part II.: * Te Kauwae-raro/ or 
'Things Terrestrial/ 279 pages. 
Cloth bound. 

Price (each Part) to Members, 
To others 

** Karotonga Records/' being extracts from 
the papers of the late W. Wyatt Gill, 
LL.D., 125 pages, map, paper cover ... 

" The Land of Tar a/- 123 pages, with map, 
paper cover 



8s. 
10s. 



3s. 
3s. 

Digitized bv LjOOQ I C 



M 



^x 



j 



Digitized by 



Google 



Digitized by 



Google 



UNMRMTY OF MOHMMN 

Hi 

3 9015 02795 1592 



Replaced with Corwrclfli Microform 
1993 



Digitized by '